motivation in software engineering a systematic literature review

popular homework editor website au

Year after year, we review dozens of reader nominations, revisit sites from past lists, consider staff favorites, and search the far-flung corners of the web for new celebration of new year essay for a varied compilation that will prove an asset to any writer, of any genre, at any experience level. This selection represents this year's creativity-centric websites for writers. These websites fuel out-of-the-box thinking and help writers awaken their choke palahnuik and literary analysis. Be sure to check out the archives for references to innovative techniques and processes from famous thinkers like Einstein and Darwin. The countless prompts, how-tos on guided imagery and creative habits, mixed-media masterpieces, and more at Creativity Portal have sparked imaginations for more than 18 years. Boost your literary credentials by submitting your best caption for the stand-alone cartoon to this weekly choke palahnuik and literary analysis from The New Yorker. The top three captions advance to a public vote, and the winners will be included in a future issue of the magazine.

Motivation in software engineering a systematic literature review do my admission essay on donald trump

Motivation in software engineering a systematic literature review

ESSAY ENGLISH UNEMPLOYMENT

If a decision cannot be reached by the researcher alone, the paper goes to arbitration. Accepted papers. The review will be conducted by one researcher Sarah. A second researcher Dorota will act as a checker by looking at all the accepted papers.

Where researchers agree, the paper will be included in the review. Where researchers disagree, the paper goes to external arbitration. Stage 1: Internal Arbitration: Researchers involved in the data extraction will try to reach an agreement on all papers whether to include or exclude. If there is still no agreement, the papers go to stage 2, external arbitration. Stage 2: External Arbitration: If the first internal arbitration fails to reach an agreement PDF s of arbitration paper s are sent to external research experts Tracy, Nathan, Helen and Hugh who, based on knowledge of our exclusion criteria, inclusion criteria and quality criteria, will make a final decision — whether to accept or reject the paper.

This is done by grouping papers by author and co- authors. Duplicate work may not be referenced by the author directly therefore papers grouped by author need to be carefully read to uncover possible duplication. Where duplication is found we include only one paper in our review that we consider to be the best quality — e.

In this way we avoid giving one finding too much prominence. The synthesis comprises qualitative lists of findings that will provide broad answers to our research questions. In order to perform sensitivity analysis we categorise the quality, population, location, year and type of study.

Population of papers Differences Similarities list list e. Students e. Computer Operators e. Novices e. When populating the results forms for each individual paper we may find further categories to investigate. Validation of review process This section explains how we validate our systematic review process - this is in four parts. The Pilot — Testing the Process a.. Three independent researchers use a subset of resources to test the process.

Problems in replicating the process are identified, process is refined accordingly This stage is completed b. Gaps in our searches are identified and search terms and resources are changed to include missing papers. Data Extraction. We test the reliability of how we extract details from accepted papers.

An independent researcher, not involved in the pilot, is given a set of accepted papers and asked to fill in the final report. The review — Testing reliability of selection d. Dorota to do a stratified sample test based on a representative sample of papers from each library that contains all the papers selected by Sarah.

Each of these four validation schemes are explained below in full. These researchers will not be involved in evaluating or arbitrating any of their own papers. The key researcher, Dr Sarah Beecham, has not published papers in this area. Should the search reveal any papers that are authored or co-authored by any of the researchers involved in this review; the author s will not be involved in the selection process.

All quality and acceptance decision made by the key researcher will be checked by a second researcher. We test that our process is fair and unbiased, replicable and open to external review. Pre-pilot: Three reviewers Dr Tracy Hall, Dorota Jagielska, and Dr Sarah Beecham are given a procedure to follow that is a step-by-step guide to extracting papers from reference databases.

These guidelines include research terms, inclusion and quality criteria. See Appendix A for the formal procedure followed. The process was trialled three times, each time the process was refined. As we were trialling the process, the actual number of accepted papers from each researcher all of whom downloaded approximately papers is not considered critical to this part of the study. The trial served to test that our inclusion criteria was understandable and our forms were workable.

Consistency results are given in Table 5. As the database used in Compendex. We therefore do not consider the face value. The final outcome is guarantee of inclusion in the not affected by these differences. Another information e. The third researcher used it moderately, but in the way intended. WIP is a temporary store, and is not relevant to the validation of the process.

When papers are further assessed they will be deleted from this library. Until all the papers in this section have been assessed, we cannot derive a final figure of accepted papers. Papers that meet the All three researchers placed papers in this section. The Venn inclusion criteria diagram shows where agreements and disagreements occur. Papers that meet inclusion criteria answers a RQ, reliable source, not a personal viewpoint.

The intersections in the Venn diagram in figure 1 show that there are 15 papers that have been selected by more than one researcher. The likelihood of the same 15 papers being selected by three researchers from by chance is extremely low2, we therefore feel that this exercise shows the process to be reliable and repeatable.

It also confirms that researchers are interpreting the inclusion criteria correctly. They serve to uncover possible misunderstandings or ambiguities in the process. For example WIP was not used by researcher number B, as this researcher was only involved in the pilot study and was not going to continue further with the study. We have not included the accepted papers library numbers since very few papers reached this stage due to lack of time.

This library and the accompanying generic accepted papers form will be fully tested by Nathan in a separate study. Nathan will be given 5 accepted papers and asked to fill in the form. Sarah will also fill in the forms and a comparison of substantive results will be made. Table 6 gives examples of missing texts identified in our pilot study and how we addressed these gaps through amending our search terms.

Outside the scope of this interdisciplinary costs and benefits of study. Work is unconnected to software enlarged jobs, Journal of Applied engineering. John Wiley and Sons. Computer Personnel 11 4 Couger, J. Adelsberger, et al. Identified in Search Colter Dorset House. DeMarco, T. Lister Academic Press, NY. Work of Fitz-enz recommended by N. Not in Compendex. Work appears to be sourced in text books We amended the search terms and tested the new terms.

An example of how we now capture the Capretz study is given below. To summarise, validating the gaps in our the pilot study revealed that we need to include more terms, and also conduct independent secondary searches on key authors, key conferences and key journals.

However, there are a few exceptions that we note below which explain why certain texts are not and will not be included in our review. Some key texts are in book form. However we eliminate the many books that have been written on motivation from our search, and focus only on resources available through indexing databases. We make this decision not only for pragmatic reasons, but also because we want to sample work that has been peer reviewed, is reliable, and reflects the thinking at time of publication.

However, we do note here that books do include relevant information, e. We will be recording secondary sources to include books in our accepted primary study paper results form. Where work is referenced in books we will endeavour to source and reference the findings in these books also. This will be a secondary phase that takes place after the systematic review of the literature available in databases.

Although we are more interested in current thinking, we need to know whether software engineer characteristics and motivation have changed over the years. There is little published in this field before this date. The exceptional key studies that pre-date the s, can be found referenced or even reproduced in more recent work.

To ensure this work, and other related work, is included in our review we have included a field in our final report form. This field prompts the researcher to record important references and secondary sources for further investigation. However, as with much published work, it is debatable if papers that show no difference ever get published or whether organisations want to publish negative results.

We realise therefore that we are creating a literature review of work that may not reflect the true state of practice. While we are interested in how groups of engineers behave and process information, we decided that this work it outside the scope of our systematic literature review of software engineer motivation.

We will include studies that reveal some general software engineer characteristics; we cannot include work on how software engineers improve productivity based on how they process information when using new or existing tools. Assessing generic studies and generic models of motivation is beyond the scope of this study.

Open Source Systems, Agile, traditional 5. When searching the databases we retrieve thousands of potentially relevant papers. From this list we need to extract those that meet our inclusion and quality criteria. The inter-rater reliability test is performed as follows. One researcher makes a search of an indexing database using search terms that cover all our research questions.

A process of selection is performed, whereby papers are checked against several criteria: Check 1. Face Value selection; Check 2. Inclusion Criteria Met; Check 3. Quality Criteria met. Only papers that go through these three checks will be included in the review. Researcher two is not told how any of these papers were categorised by researcher 1.

Search possibly key words and terms bring up some abstract, while still in rejected papers given papers that do not relate to indexing database. Inclusion — although a good abstract 2 papers from this criteria may include enough detail category are given to R2 for inclusion criteria to be as examples of papers that met are included in our accepted papers folder.

Not sure or Papers that are difficult to 5. This is a reject a potentially relevant R2 is given a temporary store — once full paper. Where it is representative 8 papers paper is found, the paper impossible to get a full from this category x will either be rejected or version of paper, the paper 5. Researcher two is given definitions of our 4 categories to include detailed notes on our research questions, exclusion criteria, and inclusion criteria.

The second researcher is then asked to place each of 3 The actual number of accepted paper is likely to be higher as there are several papers in our work in progress WIP temporary paper store. These WIP papers require more information before we can make a decision. Given the qualitative nature of most of the work in this field, we need a reliable way to report main findings from these papers.

An independent researcher researcher D is presented with five accepted papers. An example of a completed form used for this purpose is given in Appendix B. The details we place in the form do not present a summary or abstract of the paper, but provide precise concrete details of how the paper answers our research question s. It also details any secondary searches that are required.

Amendments have been made accordingly. Major amendments to the protocol have now been made in accordance with all feedback and reviews. This version 4 will be used to perform the review. Should any further changes be required we will update this protocol and change the version number accordingly. The most up-to-date version of the review will be posted on the MoMSE collaborative website Lotus Quickplace so that all researchers involved in the review have access to the current version.

DJ secondary downloaded Sarah, Hugh and Dorota joining meeting at Where of quality This will be supported by a detailed technical report that provides all the necessary transparency into the process and final reports. Making changes to the Protocol It is likely that changes to the protocol will be made when applying the procedures in new situations. Some changes will be made out of necessity, whereas other changes may be made to improve the current process.

Every change to the protocol will be recorded and the protocol updated accordingly. WebSite: gow. Thanks also to the following authors for supplying their papers and reports sometimes in draft form — they all helped to guide the development of this protocol: M. Staples and Mahmood Niazi Experiences Using Systematic Review Guidelines. Draft protocol References Capretz, L. Adelsberger Zawacki Enns, H. Ferratt, et al. Ferratt, T. Short Khalil, O. Zawacki, et al. Kitchenham, B.

MoMSE Standish Report. Wynekoop, J. Walz One Indexing service EI Compendex www. We will therefore run 5 separate searches one for each research question and limit the time intervals. Specific step by step instructions are given below. It is important to note that although we are separating the search terms into sets relating to individual research questions, IF you find a paper in ANY search that relates to another research question do not ignore it.

Because the research questions have overlapping themes, it is likely that you will get some overlapping papers in each search. If you get the same paper coming up in subsequent searches you should ignore it. Each paper should have only one results form — so please fill in all categories in the results form when you first look at the paper an example of the results form is given at the end of this document in section 3.

Implementing search dates and terms 1. For completeness, these terms are also given below. The table also contains cells for your search results you will need to complete these immediately after you perform each search. Ignore repeated papers included in earlier searches. Year: Title: What is the attraction to computing?

No Exclusion Criteria b Is study external to software engineering? No Exclusion Criteria c : Is study personal opinion piece or viewpoint? Although we are interested in general software engineer characteristics, the cognitive behaviour studies tend to be specific to certain environments and testing new methods and tools. Analysing and synthesising results from this type of work requires a separate study. Experts in this area — external to this review - will provide us with an overview of software cognitive behaviour which may prove important.

They remain here with reason for rejection entered into Endnote Decision Field as they may prove a valuable resource for other systematic reviews. Education e. IT management e. We do not exclude a paper on the basis of quality unless it is of such poor quality that we cannot interpret it. However, we will by the end of this section have an idea of the quality and whether we can generalise from the results. The Endnote Field requires you to fill in a quality score.

To get a score you need to go to the Quality Assessment and Results Form, embedded in the Endnote Field of the same name. It is a very rough measure and can only be used in the context of the type of study being performed. For example of the form see section 4. Paper is not rejected because of poor quality. We do not expect all fields in the quality section to score highly or be positive. The score represents an aggregated quality assessment. See aggregated quality assessment form in section: 4.

N a t ha n, H e le n c da t e of e nt ry Guidance: In this field list the changing status of paper — until a decision is reached. In the case of arbitration, give details of person who is arbitrating; in all other situations give name of researcher responsible for decision. Record all changes to status here with dates i. So the field might be filled with: Proceed to next stage - Sarah We retain details within primary paper for consistency and traceability.

These papers will NOT be included in the systematic review, but are stored in Related study folder. No more details are required. The results form follows on from our quality assessment is physically on the same form and is embedded into the Quality Assessment and Results field in Endnote.

External signs or outcomes of motivated engineers RQ3 4. External signs or outcomes of de-motivated software engineers RQ2 — indirect 6. Secondary studies embedded in paper that Give Reference and area of work interested in require analysis of original work before using in following up. Author direct contact If study is important, current, incomplete or suggests author is continuing to research in this area, note contact details here and what questions to ask them.

If likely, run secondary search on Author name. If you fill in any secondary source fields 9 , save file in Secondary Sources folder ensuring you fill in paper ID reference at top of form. This can include direct quotes from the paper.

Please complete all sections that are relevant, even if it appears to answer different RQ s to the one used for your search. If you fill in any primary study fields AND some secondary source fields save file in both secondary sources folder AND accepted papers folder. This completes all the possible entries you need in Endnote. Data synthesis uses: All Accepted papers: the Endnote form; the quality assessment and the results form. We have no fields in Endnote that relate to this part of the systematic review.

For how we synthesise the data see section 4. Software Engineer motivators RQ2 Recognition based on objective criteria Roles should be allowed to evolve, and then define roles and responsibilities fit roles to people, not people to roles 3. External signs or outcomes of de-motivated Sw High turnover engineers RQ2 — indirect 6. Based on hypothesis that IT staff are motivated by an enhanced working environment and sound leadership.

Secondary studies embedded in paper that Give Reference and area of work interested in following up. Author direct contact Frangos states that model is still in experimental stage. We should contact the author to find out whether this model was developed further. Do search on Frangos — may give details of development of the motivation model or related work. If you fill in secondary sources fields 9 or 10 , please save this file in Secondary Sources folder, for later follow up work.

Please ensure you fill in the paper ID reference at top of form. These search strings are copied and pasted into the database to initiate the review. To reduce the number of searches we combine RQs and perform general searches. Engine is case sensitive sees capitals as proper nouns. EI Compendex www. Compendex contains over 9 million records and references over 5, international engineering sources including journal, conference, and trade publications.

Coverage is from to present and the database is updated weekly. Researcher Name: Table 8. Google Scholar www. Can use nested Boolean search strings. You can then refine search with further Boolean search terms.. But there is a limit to how complex you can made search. I tried 2 ANDs and 3 ORs and the system froze — however there were also at least 30 terms included that add to the complexity.. The search engine is NOT case sensitive. Boolean searches possible by combining saved searches, e.

Additional comments may be provided where necessary. Is the protocol well structured, clearly written and does it cover all relevant issues? Yes Comments: It would benefit from some forward references: Section 4. Cover sheet 2. Does the sheet include the names of the reviewers and contact details of the main reviewer?

Is the title clear and does it reflect the subject of the protocol? The problem? Yes Comments: 4. The significance of the problem? Yes Comments: The authors might consider describing some of the mixed messages with references. Response: This required more background reading and explaining what some of the mixed messages are. Propose adding the following to our introduction: The literature is presenting mixed messages relating to software engineer motivation.

For example, a body of work found that programmers and analysts have lower measured needs for social interaction and higher growth needs than the general population e. They found that IT employees within the technical- professional and managerial sub-occupations of IT employees were not more motivated by achievement needs than corresponding subgroups of non-IT employees.

There is a prevalent view in the IT literature that IT employees are homogeneous in their needs suggesting that IT employees are motivated by the same employment arrangement e. Yet according to a recent study Enns et al. Current practice regarding the problem inc. Are there standard study approaches? This is possibly due to the multi-disciplinary nature of the research questions.

Theoretical studies are those where an expert makes observations and may draw on some of the motivational literature more associated with psychology and sociology and business organisation. Until the literature review is completed, it is not possible to predict whether there is a general approach to recognising SE motivation issues. Action: I have included a categorisation of study approaches in our quality assessment form to allow us to perform sensitivity analyses.

Is the background clear and understandable for the non-expert? Yes Comments: Could be more detailed. Response: By adding to what the mixed messages are item 4 and the types of studies that exist, we will be giving more detail and background to the work.

Objectives 7. Are the specific objectives and the research questions for the review clearly stated? Yes Comments: Response: None required 8. Does the proposed review address an important software engineering question?

Search strategy for identification of studies 9. Is the search strategy reasonably unbiased, comprehensive and adequate for the study question? Constructing search strings for some other sources may be different in particular ACM. When using new search engines we will check that key papers are included when making the search. We use the lookup table to copy and paste the string into the database. We do not need to amend our procedure, but add to section 4.

Information and Software Technology , 50 pp. The objective of this review is to plot the landscape of current reported knowledge in terms of what motivates developers, what de-motivates them and how existing models address motivation. Systematic reviews are well established in medical research and are used to systematically analyse the literature addressing specific research questions. Our findings suggest that Software Engineers are likely to be motivated according to three related factors: their 'characteristics' for example, their need for variety ; internal 'controls' for example, their personality and external 'moderators' for example, their career stage.

The literature indicates that de-motivated engineers may leave the organisation or take more sickleave, while motivated engineers will increase their productivity and remain longer in the organisation. Aspects of the job that motivate Software Engineers include problem solving, working to benefit others and technical challenge. Our key finding is that the published models of motivation in Software Engineering are disparate and do not reflect the complex needs of Software Engineers in their career stages, cultural and environmental settings.

It is clear that motivation is context-dependent and varies from one engineer to another. The most commonly cited motivator is the job itself, yet we found very little work on what it is about that job that Software Engineers find motivating. Furthermore, surveys are often aimed at how Software Engineers feel about 'the organisation', rather than 'the profession'. Although models of motivation in Software Engineering are reported in the literature, they do not account for the changing roles and environment in which Software Engineers operate.

Overall, our findings indicate that there is no clear understanding of the Software Engineers' job, what motivates Software Engineers, how they are motivated, or the outcome and benefits of motivating Software Engineers.

Have removed thesis freedom of movement think, that

We reviewed the extant body of research on emoji and noted the development, usage, function, and application of emoji. In this review article, we provide a systematic review of the extant body of work on emoji, reviewing how they have developed, how they are used differently, what functions they have and what research has been conducted on them in different domains. Furthermore, we summarize directions for future research on this topic.

With the widespread application of computing and the development of technology, computer mediated communication CMC is infiltrating daily life to a greater and greater extent. It has many advantages, including enhancing the continuity of individual communication Juhasz and Bradford, , improving the quality of relationships Pettigrew, ; Perry and Werner-Wilson, , and strengthening emotional communication Derks et al.

However, the lack of non-verbal cues such as facial expressions, intonation, and gestures in CMC can affect the transmission of information Archer and Akert, To address this problem, communicators have devised new non-verbal cues, such as capitalization as a substitute for shouting, multiple exclamation points for excitement, and expression symbols for facial expressions Harris and Paradice, ; Riordan and Kreuz, These expression symbols make up for the lack of non-verbal cues in CMC Tossell et al.

As a result, emoji, which are a set of expression symbols, came into being. Emoji are used more and more frequently in network communication, and the way they are used is becoming more and more diversified as well.

They not only have unique semantic and emotional features, but are also closely related to marketing, law, health care and many other areas. The research on emoji has become a hot topic in the academic field, and more and more scholars from the fields of computing, communication, marketing, behavioral science and so on are studying them. This paper reviews the developmental history and usage of emoji, details the emotional and linguistic features of emoji, summarizes the results of research on emoji in different fields, and puts forward future research directions.

We started by selecting databases. With reference to the major platforms for related publications, we chose Web of Science and Google Scholar as literature sources. Patents, news, book reviews, editorials, letters, and other literature types were excluded and the two databases were combined.

After getting rid of the duplicates, we got papers published in 78 journals and and delivered at 33 conferences. We summarized the list of journals and conferences that published articles on emoji more than once, as shown in Table 1. In this paper, the above contents were summarized and reviewed. Table 1. Statistics for articles on emoji published in journals and delivered at conferences.

In recent years, emoji have become a hot topic for research, with the volume of papers increasing gradually from and peaking at Research mainly comes from the fields of computer science and communication science.

Marketing, behavioral science, linguistics, psychology, medicine, and education are also involved. Research mostly uses empirical analysis, focusing on the diversity of individuals, cultures and platforms in the use of emoji, the attributes and characteristics of emoji, their functions in communication and the application of emoji in various research directions. Table 2 systematically summarizes the main research fields, research topics, main conclusions and research methods for emoji. Table 2.

Main research fields, research topics, main conclusions, research methods, and number of publications for emoji. In addition, a lot of researches on emoji are cross-field. For example, as emoji are platform-or system-dependent, they are often used in online communication.

Due to its visual characteristics or platform differences, there would be emotional or semantic ambiguity in communication. Many researchers from computer science try to solve this problem using a computer method and a series of algorithms or models for semantic disambiguation and sentiment analysis have been developed. Besides, the use of emoji is associated with psychological differences.

Some researchers in the field of psychology have also focused on emoji usage to search for the relationship between user's behavior and personality traits. What's more, emoji is used in marketing activities to enhance interaction and promote consumers' willingness to purchase. In order to make better use of this symbol, researchers from the field of marketing draw on relevant theories in the field of linguistics, especially in rhetoric, to enhance the appeal of emoji in marketing activities.

Emoji originated from smiley, which first evolved into emoticons, followed by emoji and stickers in recent years. Smiley first appeared in the s and is regarded as the first expression symbols. Smiley is a yellow face with two dots for eyes and a wide grin which is printed on buttons, brooches, and t-shirts.

By the early s, this symbol had become widespread, emerging as a permanent feature of western popular culture Stark and Crawford, Emoticons were introduced in and use ordinary punctuation marks from a standard computer keyboard to build up a representation of a face with a particular expression Zhou et al. They are a paralinguistic element Lee and Wagner, ; Jibril and Abdullah, often used at the end of a sentence Sakai, Prior to the existence of emoji, users of Instant Messaging IM would often use emoticons.

Like non-verbal clues in face-to-face communication, emoticons can help clarify intentions in ambiguous contexts Thompson et al. Besides, emoticons possess nonverbal communication functions. In practice, gender, and cultural differences lead to different preferences for emoticon usage Wolf, ; Jack et al. It has also been suggested that emoticons could be applied to real life, for example in fields such as emotional monitoring Carvalho et al.

The first set of emoji was released in and was created by their Japanese originator Shigetaka Kurita. Possessing similar neural responses to face-to-face communication Gantiva et al. On a social level, emoji, as a visual language, make it easier for non-English speaking nations to use English-dominated social media such as Twitter, Instagram and Facebook Boothe and Wickstrom, Emoji are widely used in instant messaging, e-mail, social networking and many other forms of CMC Dresner and Herring, As indicates, emoji fill the need for non-verbal cues in CMC to express the intentions and emotions behind information Alshenqeeti, In recent years, in order to realize the interpretability of information transmission and better express its meaning, stickers came into being Zhou et al.

Stickers can help users strategically and dynamically choose the best way to express their emotions, opinions, and intentions and to achieve communicative fluidity Lim, At the same time, stickers can be used for strategic motives such as self-presentation, impression management, establishing social existence and maintaining social status Lee et al.

Besides, responding to a partner with a combination of text and stickers can establish a high level of intimacy Wang, Smiley, emoticons, emoji, and stickers differ in form and content, and have been favored by users in different periods. Smiley, often used in advertisements and product packaging, can encourage positive moods and improve morale Stark and Crawford, Unlike emoji, emoticons, and stickers which possess a whole set of characters, smiley is a single symbol rarely used in communication.

Emoticons present facial expressions by various combinations of punctuation marks and can be used in CMC. Studies have shown that smiley and emoticons have no difference in information interpretation, but that smiley has a greater impact on individual mood than smiling emoticons Ganster et al. Emoji have come to be regarded as an advanced version of emoticons Aull, , and are superior to emoticons in terms of content richness, input speed and expressiveness Barbieri et al.

Because both act as auxiliary means of communication, emoji and emoticons are completing for similar functions. But the emergence of emoji has been proven to impact the status of emoticons to a certain extent. Compared to emoticons, users use emoji more frequently, with a more positive attitude and a deeper level of identification Prada et al. Stickers appeared recently. They are bigger, with static and animated forms, can be added or deleted emoji rely on Unicode which can't be edited.

But stickers can only be sent separately without insertion in text messages Zhou et al. Table 3 summarizes the differences between smiley, emoticons, emoji, and stickers. As emoji currently make up the most widely used and standardized symbolic language with the largest number of existing studies, this paper mainly reviews and discusses related research on emoji. As non-verbal cues in CMC, emoji are widely used in internet communication.

As of March , there were 3, emoji in Unicode, with nearly half of all text messages on Instagram containing emoji Dimson, , and 5 billion of them being used daily on Facebook. In , emoji was named the word of the year by the Oxford English Dictionary, indicating emoji's influence in online communication.

Simplicity, convenience and conduciveness to emotional expression are the main motivations attracting users to use emoji. Specifically, emoji can help users to express themselves, relax their mood Kaye et al. As contextualization cues Al Rashdi, , emoji are used in communication to promote interaction Gibson et al. In addition, emoji are also used to greet Aull, , and to maintain and enhance social relations while strengthening communication within a platform Monica Riordan, b.

However, some researchers point out that emoji may also be maliciously used for deception Njenga, In the process of using emoji, the differences between individual characteristics, platforms, cultural backgrounds, and contexts may lead to different understandings.

Emoji are also used for some specific topics, such as in sexually suggestive contexts Thomson et al. This paper systematically summarizes the differences of emoji in terms of individual diversity, cultural diversity, platform diversity, and their inefficiency in use. The use of Emoji is influenced by demographic characteristics and individual psychological characteristics.

First of all, there are significant gender differences. Although males and females understand the function of emoji similarly Herring and Dainas, , females use emoji more frequently and positively Prada et al. However, this trend varies according to communication situation.

In public communication, women are more likely to use emoji while in private communication the opposite is true Chen Z. In terms of the cognition of emoji, females perceive emoji as more familiar, clear and meaningful Rodrigues et al. Male users prefer to use the same emoji to enhance emotional expression Chen Y. When men and women use the same emoji, the recipients feel different emotions. Women who send messages containing affectionate emoji are considered more appropriate and attractive than men, and when men send messages containing less affectionate but friendly emoji messages, they are considered more appropriate and more attractive than women Butterworth et al.

The use of emoji is also affected by individual psychological differences. This has been shown in research which demonstrates a positive correlation between the frequency of emoji use among Facebook users and their extraversion and self-monitoring traits Hall and Pennington, , and a negative correlation between positive emoji use and users' emotional distress Settanni and Marengo, An emoji-based personality test indicated that the similarity score between emoji and oneself was correlated with emotional stability, extroversion and agreeableness out of the Big-Five personality traits, but not correlated with conscientiousness and openness Li et al.

Specifically, negative emojis were negatively correlated with emotional stability, while positive emoji were positively correlated with extraversion. In addition, emojis associated with blushing e. In the forum, people communicate with each other to explore the various uses and meanings of emoji. With the increase of the need to express individual diversity, people are no longer satisfied with using the existing emoji in the system, but began to create their own expressions and add more personal characteristics to emoji.

For example:. These are new symbols created by people after recombining existing emoji. Emoji use is structured by a combination of linguistic and social contexts, as well as cultural conventions Derks et al. Cultural differences have a significant impact on the use of emoji. Some specific uses of emoji are closely related to cultural background Park et al.

For example, Finnish, Indian and Pakistani users will use specific emoji according to their own culture Sadiq et al. In terms of usage behavior, following Hofstede's cultural dimension model, people in countries with high power distance and indulgence use more emoji representing negative emotions, while people in countries with high uncertainty avoidance, individualism and long-term orientation often use emoji representing positive emotions Xuan et al.

Specifically, Chinese users are more likely than Spanish users to use non-verbal cues such as emoji and emoticons to express negative emotions Cheng, Research has also found that people from Hong Kong and the US use emoji differently on user-generated restaurant reviews websites, which may reflect underlying cultural differences Chik and Vasquez, Because of the cultural differences in emoji use, an EmojiGrid was developed for cross-cultural research on food-related emotions, which reliably reflects established cultural characteristics Kaneko et al.

This difference is not only evident between countries, but also within the same country Barbieri et al. Emoji use is partly influenced by a range of national developmental indicators including life expectancy, tax rates, trade and GDP per capita. Specific language environments also affect the use of emoji.

Emoji show a high degree of context sensitivity in cross-language communication, meaning that they are exceedingly dependent on their linguistic and textual environment Vandergriff, For example, research suggests strong similarities in emoji use between Britain and America due to the fact that they both speak English, but there was less similarity when the comparison was made with other languages, such as Italian and Spanish Barbieri et al.

There are also differences in the use of emoji among specific cultural groups. One example of this is a particular style of mobile communication creatively balancing use of emoji, stickers, and text developed by some adults in rural and small towns in Southern China Zhou et al.

Research has also shown that Japanese teenagers find innovative new ways to use emoji so as to manage their relationships and express themselves aesthetically in a subculturally specific way Sugiyama, The use of Emoji is also related to interpersonal relationships Gaspar et al. The more polite and distant the conversation between people, the more abstract, geometric and static the emoji will become.

On the contrary, more specific and vivid emoji is used in groups where participants are more sympathetic to a particular topic, more companionate and more intimate Lin and Chen, Platform diversity is one of the important factors affecting emoji use. The presentation style of emoji on different operating system platforms and the architectural specifications of different network platforms will affect users' preferences for emoji.

Although emoji use Unicode, the presentation style of emoji in IOS, Android, Microsoft and other systems is different due to the influence of different developers as shown in Table 4. Studies have found that emoji on the IOS platform are more aesthetically attractive, familiar, clear and meaningful than those on the Android platform Rodrigues et al.

This difference in platform display will lead to misunderstanding and divergence in terms of emoji's emotional and semantic interpretation during cross-platform use Tigwell and Flatla, In addition, researchers have studied different network platforms such as Twitter, Facebook and Instagram, and found that users of different platforms have their unique preferences when using emoji.

The most popular emoji in one platform may not be popular on other platforms. For example, users tend to use emoji more frequently and positively on Facebook than on Twitter Tauch and Kanjo, At the same time, some researchers focus on more marginal community platforms like Gab. In the face of the same event, Gab users tend to publish positive emoji to express irony in text with negative connotations, while Twitter users tend to use emoji to express suspicion Mahajan and Shaikh, However, some researchers believe that the use of emoji is generally consistent on all platforms, except for e-mail, which is not suitable for using emoji Kaye et al.

Emoji can help users to convey feelings and understand the meaning of a text, but the use of emoji also brings ambiguities in the interpretation of communication, resulting in inefficiency. Although emoji have visual similarity, their interpretation is influenced by cultural background, technical differences and their own visual characteristics Bich-Carriere, The specific meanings that users want to express by emoji may be different from their official definitions, resulting in different interpretations of the same emoji Miller et al.

In this case, it is difficult for the two sides to understand each other, which reduces the efficiency of communication. Berengueres and Castro found that there are differences in understanding negative emoji. Research done by Riordan a shows that the degree of misunderstanding of facial emoji is higher than that of non-facial emoji, but that both are related to the degree of information ambiguity. When used across platforms, the differences in how people interpret emoji emotionally and semantically will increase because of platform display differences Miller et al.

The difference in how emoji are understood results in inefficiency in communication, leads to the interruption of discourse and destroys interpersonal relationships Tigwell and Flatla, As an important visual symbol in computer-mediated-communication, emoji can express various content, including people, animals, food, activities. Emoji can be used both as an independent language and a non-verbal cue to convey meanings, which is the semantic function of emoji. In addition, emoji also have emotional functions.

We have summarized them in Table 6. Because they are non-verbal cues with rich emotional meanings, emoji are an important medium for interaction and emotional communication on the Internet. Jaeger and Ares analyzed 33 facial emojis and found that most emoji can express one or more emotions.

The rich emotional meaning of emoji makes them a key area for researchers who analyze their emotions and develop emoji emotional lexicons. By artificial annotating, Petra et al. Due to the subjectivity of human annotating, some researchers have proposed the automatic construction of emoji lexicons.

Fernandez-Gavilanes et al. Because of their rich emotional meanings, emoji are often used to express emotions in online communication. In general, users tend to use emoji in positive messages and to use them less in sad or angry messages Cheng, Different emoji affect people's attention and responses in divergent ways Hjartstrom et al.

Although both facial and non-facial emoji can express emotions Riordan, a , facial emoji outperform non-facial emoji Jaeger et al. Using non-facial emoji can bring about positive emotions, especially joy, but it can't change the valence of the message Riordan, b.

In addition to expressing emotions, emoji are also used to convey semantic meanings in communication Na'aman et al. They can play the role of non-verbal cues to help understand the overall meaning of messages in CMC Walther and D'Addario, ; Jibril and Abdullah, There has been a lot of discussion about whether emoji could become an independent language. In addition, due to the diversity and similarity of emoji semantics, many researchers from the field of computing pay attention to the word sense disambiguation task of emoji.

Some research suggests that emoji form an independent language. Compared with plain text, emoji are richer in semantic meaning Ai et al. At the application level, Khandekar et al. However, some researchers suggest that emoji can't be used as an independent language. Lee et al. Alshenqeeti argues that emoji is essentially a form of visual paralanguage. Furthermore, emoji tend to be text-related and rarely used independently.

Emoji need to be integrated with the text in order to form a complete meaning Zhou et al. In practice, users tend to use emoji as a supplement to text Ai et al. The meaning of emoji varies according to specific context Gawne and McCulloch, Their diversity of semantics and flexibility of interpretation may lead to ambiguity when using them Jaeger et al.

Therefore, a lot of research focuses on the word sense disambiguation task of emoji. Wijeratne et al. Barbieri et al. Emoji have both emotional and semantic functions and are popular in computer-mediated-communication. Researchers from different fields have studied emoji from different perspectives, including computer science, communication, marketing, behavioral science, linguistics, psychology, medicine, and education.

Research in the field has focused on using emoji for emotional analysis of UGC data, the conversion of emoji to other expression modality, and using emoji for optimizing computer systems. With the significant growth of UGC data on the Internet, sentiment analysis which aims at changing this data into valuable asset for decision making, has become increasingly important Al-Azani et al.

As emoji are widely used in expressing emotions, they have become an effective means of sentiment analysis Hogenboom et al. A number of studies have confirmed the effective performance of emoji in sentiment analysis Sari et al.

Besides, emoji-based sentiment analysis is language-independent and exhibits cross-language validity Guthier et al. However, other studies have shown that using emoji in sentiment analysis leads to higher emotional scores, and that this effect is more pronounced in positive comments Ayvaz and Shiha, Many studies have provided algorithms and models for emoji-based sentiment analysis, which mainly uses two kinds of techniques, sentiment lexicon, and machine learning. The sentiment lexicon approach focuses on building an emoji emotional lexicon to support text sentiment analysis.

By human annotating, Petra et al. But because there are so many emoji, some researchers have come up with ways to build emoji dictionaries automatically. Jiang et al. Kimura and Katsurai assigned multi-dimensional emotional vectors to emoji by calculating the co-occurrence frequency of emoji and emotional words in WordNet-Affect.

Aoki and Uchida have also automatically generated emoji vectors based on the relationship between emotional words and emoji. By using the Word2Vec clustering method, Mayank et al. The machine learning method refers to train sentiment classifiers based on a corpus in order to analyze the sentiments of text Wang et al. Machine learning can be divided into supervised learning and unsupervised learning.

They are different in that the former needs a human annotated corpus while the latter doesn't. The effectiveness of using emoji as a way of training classifiers has been proven Hallsmar and Palm, and furthermore it has been shown that emoji outperform emoticons Redmond et al. A lot of research has focused on unsupervised learning Li et al. Chen Y. Wang et al. Some research has focused on the ironic features of emoji and developed an irony detection model for emoji in order to improve the accuracy of sentiment analysis of tweets Reyes et al.

The visual features and Unicode basis of emoji make them an independent expressive modality that is different from text and pictures Cappallo et al. A lot of research focuses on conversion between emoji and other modalities such as text, picture and video. For example, Emoji2Video offers a way to search for videos using emoji Cappallo et al. Later research has focused on the shift from other modalities to emoji.

Because of the correlation between emotional categories in text and users' emoji selections, Hayati and Muis and Zanzotto and Santilli proposed two different ways to predict emoji based on text. Kim et al. The practice of text-based emoji prediction has also been validated in other languages, such as Hebrew Liebeskind et al. Emoji have played a role in improving the performance of computer hardware and software.

For example, emoji can be used to achieve diverse in-car interaction design. In order to optimize the functions of the central rear-view mirror, researchers suggest that passengers emotions can be fed back to the driver through emoji and other elements, which can enhance mutual understanding between driver and back-seat passenger Chao et al.

Furthermore, emoji can also be applied in the area of password security. Kraus et al. Compared with the Standard PIN Personal Identification Number input, a password containing emoji is easier to remember and, thus, emoji-based authentication is a practical alternative to traditional PIN authentication.

In the field of communication, research on emoji mainly focuses on two aspects: one is emoji's emotional and linguistic functions in CMC, the other is how different factors, such as individual characteristics, cultural background and system platform, influence users' preferences for emoji use. For example, Jaeger and Ares analyzed the emotional attributes expressed by 33 facial emojis, and found that most emoji contained one or more emotional meanings. Based on their emotional distribution, Petra et al.

Similar studies have found that users tend to use more emoji more in positive messages than negative messages. Both facial and non-facial emoji exhibit a great deal of ability when it comes to expressing emotions Herring and Dainas, ; Jaeger et al. At the same time, different combinations of emojis can enrich the meanings of emotional expression. More research in this area is referred to in section The Emotional Functions of Emoji of this paper.

Individual use of emoji is influenced by many factors. The existing research can be divided into three categories: individual characteristics, cultural background and system platform. First, the use of emoji is strongly influenced by demographic characteristics such as gender and age of users. Women use them more frequently and men use them more abundantly Tossell et al.

Women use emoji more in public communication, but less in private communication Li et al. In terms of social cognition, emoji with stronger emotional meanings are considered more appropriate and lovely for women than for men, while emoji with weaker emotional meanings but friendlier meanings are considered more appropriate for men Derks et al.

Secondly, the use of emoji is closely related to the user's cultural background. Users in different countries will use emoji with specific national or ethnic meanings Gaspar et al. Users in different countries tend to use emoji differently. Chinese people use emoji and emoticons more often than Spaniards Lin and Chen, Emoji show a high degree of contextual sensitivity and different language types influence the use of emoji.

For example, the use of emoji displays a strong correlation among English-speaking countries, while displaying lower correlation among other languages such as Italian and Spanish Barbieri et al. Finally, different system platforms also lead to differences in emoji usage. Although emoji has a Unicode in the operating system platform, users show emoji differently in IOS, Android and Microsoft operating systems due to the limitation of these software's developmental compatibility Cramer et al.

Different social networking platforms such as Twitter, Facebook, Gab and Instagram also have their own particular patterns of emoji usage. For example, users tend to use emoji more frequently and positively on Twitter than on Facebook Hall and Pennington, Gab users tend to use positive emoji to express negative emotions, thus showing irony Settanni and Marengo, More research is referred to in section Diversity of Emoji Use of this paper.

Due to their visual and emotional attributes, emoji can be used in marketing activities. Emoji play an important role in attracting attention, stimulating social interactions and enhancing the experience of consumers, along with their willingness to purchase Das et al. So it is hardly surprising that emoji are frequently used in consumer interactions Lee et al. Furthermore emoji are also used to depict consumer emotions Li et al.

Their dominance in emotional expression makes them an effective tool to measure user's emotions. Textual paralanguages like emoticons and emoji, can influence the cognition and behavior of consumers in marketing activities Luangrath et al. It has been found that using emoji can enhance the explanatory power, attractiveness, creativity and innovation of marketing activity. With the introduction of emoji in online marketing, more young people are attracted Yakin and Eru, Ge and Gretzel indicate that social media influencers people who take on the dual roles of marketer and active user of social media can initiate online interaction by presenting emoji individually or in combination, which can attract consumers to participate in interactions.

Emoji can also be a way of reflecting consumers' emotions, describing user's profiles Moreno-Sandoval et al. It has been found that gender, age and frequency of usage do not affect consumers' ability to describe and distinguish stimuli with emoji Jaeger et al.

In addition, emoji and emoticons are considered simple and intuitive ways to express food-related emotions Vidal et al. Marketers use emoji questionnaires as a common tool to measure user's emotions Jaeger et al. However, some researchers point out that although emoji show more discriminability and simplicity than emotional words in emotional measurement, their multiple meanings could pose a barrier to the survey.

Therefore, emoji questionnaires can't directly replace the existing text-based forms of sentiment survey directly. They can, however, act as a complement to the current form Jaeger et al. In the field of behavioral science, research on emoji focuses on three aspects: motivation, preference and influencing factors.

There has been abundant research focusing on the motivations behind emoji usage. This research has found that emoji are used for managing and maintaining interpersonal relationships Chairunnisa and Benedictus, ; Riordan, b ; Albawardi, , expressing oneself Kaye et al.

As a contextual cue, emoji can help users establish an emotional tone, reduce the ambiguity of semantic expression and improve appropriateness relative to context Kaye et al. There are two aspects of emoji usage preference. One is users' selection of emoji content and the other is the degree to which there is a match between emotions expressed by emoji and real sentiments. For example, users in different countries introduce elements which are representative of their countries into emoji Sadiq et al.

In the field of linguistics, research focuses on the pragmatic functions of emoji and the possibility that they could become an independent language. Emoji have been identified as having semantic properties, and can be used both as an independent language and as a component of a paralanguage providing users with a means of communication and promoting speech acts and interaction Jibril and Abdullah, ; Alshenqeeti, ; Na'aman et al. There are pros and cons regarding whether emoji can become an independent language.

An application was developed to verify the possibility of emoji-first communication Khandekar et al. Other researchers think emoji can't be regarded as an independent language because their meaning largely depends on surrounding text, and only when they are combined with the text can complete semantics be expressed Zhou et al.

Studies in this field mainly focus on two aspects. One is the relationship between individual psychological characteristics and emoji usage, and the other is the introduction of emoji into the scale design and the implementation of new psychological measurement tools. Emoji usage was found to be closely related to some psychological traits such as the big five personality traits, self-monitoring, emotional stress, and others Derks et al.

For example, research has shown that frequency of emoji use correlates with emotional stability, extroversion and agreeableness in the big five personality traits, but not with conscientiousness and openness Li et al.

At the same time, some studies have attempted to introduce emoji into psychometric scales and have achieved good results in actual measurements Marengo et al. In the field of medicine and public health, studies on emoji mainly focus on correcting personal behavior and improving doctor-patient communication.

Emoji can be used to guide people's behavior regarding health, and it has been shown that using emoji can reinforce correct behavior when it comes to hand hygiene monitoring Gaube et al. Furthermore, using emoji can improve communication between doctors and patients and also enhance patients' abilities to manage their own health Balas et al. Some researchers suggest developing a set of emoji specifically to be used for patient care, which could help patients better understand and communicate the challenges they face in health management Skiba, In addition, emoji can be used for the identification and prediction of mental illness due to their strength in emotional expression.

Marengo et al. In the field of education, research focuses on the impact of emoji on learning efficiency. It has been found that the use of emoji in classroom activities will help students better understand what they have learned Brody and Caldwell, , especially in computer-mediated teaching online learning Dunlap et al. Emoji can help young children understand abstract concepts such as security, interpersonal management and emotions and also improve their ability to express themselves Fane, ; Fane et al.

Understanding users' real emotions when they use emoji is important for future research. At present, it is difficult to accurately measure participants' true reactions through self-reporting. Categorizing emotions by amassing a corpus using big data is unable to depict users' complex emotions such as are expressed by emoji at a more detailed level, for example emotions such as shame, anger and so on.

Therefore, we hold the opinion that in the future, researchers can use some psychological methods in the corpus test to measure the physiological indexes of participants with professional equipment such as nuclear magnetic resonance, electroencephalography and multipurpose polygraphs to depict users' real emotions more accurately.

Future research could also benefit from a more qualitative approach, such as interviews and case studies to learn about emoji use in the context of real-world communication. In practice, some researchers suggest that video and screen shots can be used in concrete operations to observe and record users' choices of emoji during communication Gibson et al. We believe that observing whether users' actual facial expressions differ from their selected emoji emotionally in communication can help researchers understand users' psychological mechanism in communication.

At present, research focuses on the description of users' preference for emoji, but fails to go deeply into the underlying reasons. Users' preferences for emoji are influenced by many factors such as contextual information, interpersonal relationships, familiarity with emoji and personal interpretations other than official definitions, which are all worthwhile factors to explore.

The emergence and widespread use of stickers has impacted the status of emoji, and some research has begun to improve the user experience of stickers Shi et al. Whether stickers will replace emoji is an interesting topic for researchers. Under the impact of stickers, how to further enhance emoji's performance in emotion and semantic expression and improve user experience is also a direction worth exploring.

As part of popular culture, the development and use of emoji reflects specific political and cultural characteristics. Many researchers have interpreted emoji's social influence from different perspectives. For example, some uncivilized use of emoji can harm public consciousness, a point which is not yet appreciated by the public Zerkina et al.

Other researchers believe that the popularity of emoji reflects multicultural communication and cultural globalization Skiba, , and that there is some unconscious power behind the use of non-verbal cues like emoji Elder, , which strengthen the inequality and exploitation of our social system Stark and Crawford, For example, Leslie argues that the quantitative use of emoji in the workplace such as the use of emoji to give ratings has turned the employee into something like an on-the-shelf item in a digital economy warehouse, affecting their freedom.

The democratization of emoji selection and Unicode should also be discussed. Emoji of different skin colors have been introduced to address the lack of racial representation Sweeney and Whaley, Therefore, future research can explore the deeper meaning of emoji use from different perspectives, especially the links between emoji use and political movements, subcultural groups, and social inequality.

This paper systematically reviews related research on emoji, aiming to provide a global perspective and clues for researchers interested in emoji. This paper summarizes the developmental process, usage features, functional attributes, and fields of research related to emoji. Emoji developed from emoticons, and have both emotional and semantic functions. The use of emoji is influenced by and varies according to factors such as individual circumstances, culture, and platforms.

Ambiguity and misunderstanding may occur in different situations and cultural backgrounds. From the perspective of many fields communication, computing, behavioral science, marketing, and education , this paper comprehensively combs the research topics, methods and tools used in studies related to emoji, systematically summarizes the research status of emoji in various fields, and puts forward some new perspectives for future emoji research such as emotional association, use preference, new modalities and impacts on society.

The datasets generated for this study are available on request to the corresponding author. QB contributed conception and design of the study. All authors contributed to manuscript revision, read, and approved the submitted version. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Ai, W. Google Scholar. Al Rashdi, F. Functions of emojis in WhatsApp interaction among Omanis. Discourse Context Media 26, — Al-Azani, S. As with digital technologies in general, digital data do not offer a neat technical fix to education dilemmas — no matter how compelling the output might be. We should not strive for what is technically possible, but always ask ourselves what makes pedagogical sense. In China, systems are already being used to monitor student participation and expressions via face recognition in classrooms so called Intelligent Classroom Behavior Management System, Smart Campus Footnote 8 and display them to the teacher on a dashboard.

This is an example of educational surveillance, and it is highly questionable whether such systems provide real added value for a good teacher who should be able to capture the dynamics in a learning group online and in an on-campus setting and respond empathically and in a pedagogically meaningful way.

In this sense, it is crucial to adopt an ethics of care Prinsloo, to start thinking on how we are exploring the potential of algorithmic decision-making systems that are embedded in AIEd applications. Even the smartest AI systems can make very stupid mistakes. The new UNESCO report on challenges and opportunities of AIEd for sustainable development deals with various areas, all of which have an important pedagogical, social and ethical dimension, e.

That being said, a stunning result of this review is the dramatic lack of critical reflection of the pedagogical and ethical implications as well as risks of implementing AI applications in higher education. More research is needed from educators and learning designers on how to integrate AI applications throughout the student lifecycle, to harness the enormous opportunities that they afford for creating intelligent learning and teaching systems.

The low presence of authors affiliated with Education departments identified in our systematic review is evidence of the need for educational perspectives on these technological developments. The lack of theory might be a syndrome within the field of educational technology in general. This kind of research is now possible through the growth of computing power and the vast availability of big digital student data.

However, at this stage, there is very little evidence for the advancement of pedagogical and psychological learning theories related to AI driven educational technology. It is an important implication of this systematic review, that researchers are encouraged to be explicit about the theories that underpin empirical studies about the development and implementation of AIEd projects, in order to expand research to a broader level, helping us to understand the reasons and mechanisms behind this dynamic development that will have an enormous impact on higher education institutions in the various areas we have covered in this review.

It is beyond the scope of this article to discuss the various machine learning methods for classification and prediction. Readers are therefore encouraged to refer to the literature referenced in the articles that are included in this review e. Acikkar, M. Support vector machines for predicting the admission decision of a candidate to the School of Physical Education and Sports at Cukurova University. Article Google Scholar. Adamson, D. Towards an agile approach to adapting dynamic collaboration support to student needs.

International Journal of Artificial Intelligence in Education , 24 1 , 92— Agaoglu, M. Predicting instructor performance using data mining techniques in higher education. IEEE Access , 4 , — Ahmad, H. Lecturer performance analysis using multiple classifiers. Journal of Computer Science , 12 5 , — Alfarsi, G. A rule-based system for advising undergraduate students.

Alkhasawneh, R. Developing a hybrid model to predict student first year retention in STEM disciplines using machine learning techniques. Google Scholar. Aluko, R. Predicting the academic success of architecture students by pre-enrolment requirement: Using machine-learning techniques.

Construction Economics and Building , 16 4 , 86— Aluthman, E. International Journal of English Linguistics , 6 5 , 54— Amigud, A. Using learning analytics for preserving academic integrity. Andris, C. Support vector machine for spatial variation. Transactions in GIS , 17 1 , 41— Aparicio, F.

Perceptions of the use of intelligent information access systems in university level active learning activities among teachers of biomedical subjects. International Journal of Medical Informatics , December , 21— Machine learning methods in predicting the student academic motivation. Croatian Operational Research Review , 8 2 , — Kuram ve Uygulamada Egitim Bilimleri , 16 3 , — Bakeman, R. Observing interaction - an introduction to sequential analysis.

Cambridge: Cambridge University Press. Book Google Scholar. Baker, R. Stupid Tutoring Systems, Intelligent Humans. International Journal of Artificial Intelligence in Education , 26 2 , — Baker, T. Educ-AI-tion rebooted? Exploring the future of artificial intelligence in schools and colleges.

Barker, T. An automated feedback system based on adaptive testing: Extending the model. International Journal of Emerging Technologies in Learning , 5 2 , 11— An automated individual feedback and marking system: An empirical study. Personalisation in educational technology: The absence of underlying pedagogies.

Ben-Zvi, T. Measuring the perceived effectiveness of decision support systems and their impact on performance. Decision Support Systems , 54 1 , — Biletska, O. A semantic approach to expert system for e-assessment of credentials and competencies. Expert Systems with Applications , 37 10 , — Blikstein, P.

Programming pluralism: Using learning analytics to detect patterns in the learning of computer programming. Journal of the Learning Sciences , 23 4 , — Brunton, J. Information management in systematic reviews. Gough, S. Thomas Eds. London: SAGE. Calvo, R. Collaborative writing support tools on the cloud. Camacho, D. Towards an automatic monitoring for higher education learning design. International Journal of Metadata, Semantics and Ontologies , 2 1 , 1. Casamayor, A. Intelligent assistance for teachers in collaborative e-learning environments.

More than tools? Making sense of he ongoing digitizations of higher education. Chaudhri, V. Inquire biology: A textbook that answers questions. AI Magazine , 34 3 , 55— Chen, J. Training neural networks to predict student academic performance: A comparison of cuckoo search and gravitational search algorithms.

International Journal of Computational Intelligence and Applications , 13 1. Chi, M. Empirically evaluating the application of reinforcement learning to the induction of effective and adaptive pedagogical strategies. User Modeling and User-Adapted Interaction , 21 1 , — Chodorow, M.

The utility of article and preposition error correction systems for English language learners: Feedback and assessment. Language Testing , 27 3 , — Chou, C. Complementary machine intelligence and human intelligence in virtual teaching assistant for tutoring program tracing. Cobos, C. A hybrid system of pedagogical pattern recommendations based on singular value decomposition and variable data attributes.

Information Processing and Management , 49 3 , — Cohen, J. A coefficient of agreement for nominal scales. Educational and Psychological Measurement , 20 , 37— Contact North. Ten facts about artificial intelligence in teaching and learning. Crown, S. Anne G. Neering: Interactive chatbot to engage and motivate engineering students. Computers in Education Journal , 21 2 , 24— DeCarlo, P.

The design and development of an expert system prototype for enhancing exam quality. International Journal of Advanced Corporate Learning , 3 3 , 10— Delen, D. A comparative analysis of machine learning techniques for student retention management. Decision Support Systems , 49 4 , — Predicting student attrition with data mining methods. Dikli, S. The nature of automated essay scoring feedback.

Dobre, I. Mediterranean Journal of Social Sciences , 5 19 , 84— Dodigovic, M. Artificial intelligence and second language learning: An efficient approach to error remediation. Language Awareness , 16 2 , 99— Duarte, M. An intelligent universal virtual laboratory UVL. Duffy, M. Motivation matters: Interactions between achievement goals and agent scaffolding for self-regulated learning within an intelligent tutoring system.

Computers in Human Behavior , 52 , — Duzhin, F. Machine learning-based app for self-evaluation of teacher-specific instructional style and tools. Education Sciences , 8 1. Easterday, M. The logic of design research. Learning: Research and Practice , 4 2 , — Horizon report: higher education edition. Feghali, T. A web-based decision support tool for academic advising.

Feng, S. Research on data mining in university admissions decision-making. International Journal of Advancements in Computing Technology , 3 6 , — Fleiss, J. Statistical methods for rates and proportions. New York: Wiley. Garcia-Gorrostieta, J. Automatic argument assessment of final project reports of computer engineering students.

Computer Applications in Engineering Education, 26 5 , — Ge, C. Application of grey forecasting model based on improved residual correction in the cost estimation of university education. International Journal of Emerging Technologies in Learning , 10 8 , 30— Gierl, M. Automated essay scoring and the future of educational assessment in medical education. Medical Education , 48 10 , — Gough, D. An introduction to systematic reviews , 2nd ed.

Gutierrez, G. Mining: Students comments about teacher performance assessment using machine learning algorithms. Hall Jr. Customized content delivery for graduate management education: Application to business statistics. Journal of Statistics Education , 16 3. Haugeland, J. Artificial intelligence: The very idea.

Cambridge, Mass. Hew, K. British Journal of Educational Technology , 50 3 , — Hinojo-Lucena, F. Artificial intelligence in higher education: A bibliometric study on its impact in the scientific literature. Education Sciences , 9 1 , Hoffait, A. Early detection of university students with potential difficulties. Decision Support Systems , , 1— Hooshyar, D. A flowchart-based intelligent tutoring system for improving problem-solving skills of novice programmers.

Journal of Computer Assisted Learning , 31 4 , — Howard, C. Shifting the load: A peer dialogue agent that encourages its human collaborator to contribute more to problem solving. International Journal of Artificial Intelligence in Education , 27 1 , — Howard, E. Contrasting prediction methods for early warning systems at undergraduate level.

Internet and Higher Education , 37 , 66— Huang, C. Developing an intelligent diagnosis and assessment e-Learning tool for introductory programming. Huang, J. The research and design of web-based intelligent tutoring system. International Journal of Multimedia and Ubiquitous Engineering , 11 6 , — Huang, S. Effects of using artificial intelligence teaching system for environmental education on environmental knowledge and attitude. Hussain, M. Student engagement predictions in an e-Learning system and their impact on student course assessment scores.

Computational Intelligence and Neuroscience. Iglesias, A. Reinforcement learning of pedagogical policies in adaptive and intelligent educational systems. Jackson, M. Is intelligent online tutoring software useful in refreshing financial accounting knowledge? Jain, G. Jeschike, M. Equipping virtual laboratories with intelligent training scenarios.

Jia, J. AI Magazine , 30 2 , 59— Jonassen, D. Constructivism and computer-mediated communication in distance education. American Journal of Distance Education , 9 2 , 7— Kalz, M. A model for new linkages for prior learning assessment. Campus-Wide Information Systems , 25 4 , — Using an e-Learning system with integrated concept maps to improve conceptual understanding. International Journal of Instructional Media , 37 2 , — Kaplan, A. On the interpretations, illustrations, and implications of artificial intelligence.

Business Horizons , 62 1 , 15— Kardan, A. A decision support system for course offering in online higher education institutes. International Journal of Computational Intelligence Systems , 6 5 , — Prediction of student course selection in online higher education institutes using neural network.

Computers and Education , 65 , 1— Kose, U. International Journal of Engineering Education , 32 1, A , — Li, X. Intelligent agent-supported online education. Decision Sciences Journal of Innovative Education , 5 2 , — Lo, J.

Computers and Education , 58 1 , — Lodhi, P. StuA: An intelligent student assistant. Luckin, R. Intelligence unleashed - an argument for AI in education. Ma, H. McNamara, D. A hierarchical classification approach to automated essay scoring.

Assessing Writing , 23 , 35— Misiejuk, K. State of the field report on learning analytics. SLATE report —2. Miwa, K. An intelligent tutoring system with variable levels of instructional support for instructing natural deduction. Transactions of the Japanese Society for Artificial Intelligence , 29 1 , — Moher, D. BMJ , , b Nehm, R. Transforming biology assessment with machine learning: Automated scoring of written evolutionary explanations.

Journal of Science Education and Technology , 21 1 , — Neumann, W. Social research methods: Qualitative and quantitative approaches. Boston: Pearson. Ng, S. Design of an agent-based academic information system for effective education management.

Information Technology Journal , 10 9 , — Nguyen, J. A linguistic multi-criteria decision-aiding system to support university career services. Applied Soft Computing Journal , 67 , — Nicholas, D. Peer review: still king in the digital age. Learned Publishing , 28 1 , 15— Oztekin, A. A hybrid data analytic approach to predict college graduation status and its determinative factors. Industrial Management and Data Systems , 8 , — Ozturk, Z. Sentiment analysis: An application to Anadolu University.

Acta Physica Polonica A , 3 , — Palocsay, S. A study of the effectiveness of web-based homework in teaching undergraduate business statistics. Decision Sciences Journal of Innovative Education , 6 2 , — Paquette, L. Designing a knowledge representation approach for the generation of pedagogical interventions by MTTs.

Payne, V. Effect of a limited-enforcement intelligent tutoring system in dermatopathology on student errors, goals and solution paths. Artificial Intelligence in Medicine , 47 3 , — Artificial intelligence in education: Challenges and opportunities for sustainable development. Perez, S. Identifying productive inquiry in virtual labs using sequence mining.

Baker, X. Hu, M. Chapter Google Scholar. Perin, D. Assessing text-based writing of low-skilled college students. International Journal of Artificial Intelligence in Education , 28 1 , 56— Petticrew, M. Systematic reviews in the social sciences: A practical guide. Malden; Oxford: Blackwell Pub. Phani Krishna, K. Student information system and performance retrieval through dashboard. Popenici, S. Exploring the impact of artificial intelligence on teaching and learning in higher education.

Research and Practice in Technology Enhanced Learning. Prinsloo, P. E-Learning and Digital Media , 14 3 , — Quixal, M. How can writing tasks be characterized in a way serving pedagogical goals and automatic analysis needs? Calico Journal , 33 1 , 19— Raju, D. Exploring student characteristics of retention that lead to graduation in higher education using data mining models.

Journal of Educational Computing Research , 56 1 , 23— Ray, R. Teaching college level content and reading comprehension skills simultaneously via an artificially intelligent adaptive computerized instructional system. Reid, J. Managing learner support. Lockwood Ed.

London: Routledge. Rovira, S. Data-driven system to predict academic grades and dropout. PLoS One , 12 2 , 1— Russel, S. Artificial intelligence - a modern approach. New Jersey: Pearson Education. Salmon, G. E-moderating - the key to teaching and learning online , 1st ed. Samarakou, M. An advanced eLearning environment developed for engineering learners.

International Journal of Emerging Technologies in Learning , 10 3 , 22— Sanchez, E. Future Internet , 8 2. Schiaffino, S. Sebastian, J. Changing stigmatizing attitudes to mental health via education and contact with embodied conversational agents. Computers in Human Behavior , 73 , — Selwyn, N. Is technology good for education? Shen, V. Intelligent multiagent tutoring system in artificial intelligence. International Journal of Engineering Education , 27 2 , — Measures of diagnostic accuracy: Basic definitions.

Smith, R. Peer review: a flawed process at the heart of science and journals. Journal of the Royal Society of Medicine , 99 , — Spikol, D. Supervised machine learning in multimodal learning analytics for estimating success in project-based learning. Journal of Computer Assisted Learning , 34 4 , — Sreenivasa Rao, K. Educational data mining for student placement prediction using machine learning algorithms.

Steenbergen-Hu, S. Journal of Educational Psychology , 2 , — Sultana, S. Predicting performance of electrical engineering students using cognitive and non-cognitive features for identification of potential dropouts. International Journal of Electrical Engineering Education , 54 2 , — Tai, D. Effective e-learning recommendation system based on self-organizing maps and association mining.

Electronic Library , 26 3 , — Tegmark, M. Life 3. London: Penguin Books. Teshnizi, S. Acta Informatica Medica, 23 5 , Thatcher, S. The use of artificial intelligence in the learning of flight crew situation awareness in an undergraduate aviation programme.

Umarani, S. Speech based question recognition of interactive ubiquitous teaching robot using supervised classifier. Umer, R. On predicting academic performance with process mining in learning analytics. Journal of Research in Innovative Teaching , 10 2 , —

Think, that free essay fences august wilson this remarkable

Researchers from different fields have studied emoji from different perspectives, including computer science, communication, marketing, behavioral science, linguistics, psychology, medicine, and education. Research in the field has focused on using emoji for emotional analysis of UGC data, the conversion of emoji to other expression modality, and using emoji for optimizing computer systems. With the significant growth of UGC data on the Internet, sentiment analysis which aims at changing this data into valuable asset for decision making, has become increasingly important Al-Azani et al.

As emoji are widely used in expressing emotions, they have become an effective means of sentiment analysis Hogenboom et al. A number of studies have confirmed the effective performance of emoji in sentiment analysis Sari et al. Besides, emoji-based sentiment analysis is language-independent and exhibits cross-language validity Guthier et al.

However, other studies have shown that using emoji in sentiment analysis leads to higher emotional scores, and that this effect is more pronounced in positive comments Ayvaz and Shiha, Many studies have provided algorithms and models for emoji-based sentiment analysis, which mainly uses two kinds of techniques, sentiment lexicon, and machine learning.

The sentiment lexicon approach focuses on building an emoji emotional lexicon to support text sentiment analysis. By human annotating, Petra et al. But because there are so many emoji, some researchers have come up with ways to build emoji dictionaries automatically. Jiang et al. Kimura and Katsurai assigned multi-dimensional emotional vectors to emoji by calculating the co-occurrence frequency of emoji and emotional words in WordNet-Affect. Aoki and Uchida have also automatically generated emoji vectors based on the relationship between emotional words and emoji.

By using the Word2Vec clustering method, Mayank et al. The machine learning method refers to train sentiment classifiers based on a corpus in order to analyze the sentiments of text Wang et al. Machine learning can be divided into supervised learning and unsupervised learning. They are different in that the former needs a human annotated corpus while the latter doesn't. The effectiveness of using emoji as a way of training classifiers has been proven Hallsmar and Palm, and furthermore it has been shown that emoji outperform emoticons Redmond et al.

A lot of research has focused on unsupervised learning Li et al. Chen Y. Wang et al. Some research has focused on the ironic features of emoji and developed an irony detection model for emoji in order to improve the accuracy of sentiment analysis of tweets Reyes et al. The visual features and Unicode basis of emoji make them an independent expressive modality that is different from text and pictures Cappallo et al.

A lot of research focuses on conversion between emoji and other modalities such as text, picture and video. For example, Emoji2Video offers a way to search for videos using emoji Cappallo et al. Later research has focused on the shift from other modalities to emoji. Because of the correlation between emotional categories in text and users' emoji selections, Hayati and Muis and Zanzotto and Santilli proposed two different ways to predict emoji based on text.

Kim et al. The practice of text-based emoji prediction has also been validated in other languages, such as Hebrew Liebeskind et al. Emoji have played a role in improving the performance of computer hardware and software. For example, emoji can be used to achieve diverse in-car interaction design. In order to optimize the functions of the central rear-view mirror, researchers suggest that passengers emotions can be fed back to the driver through emoji and other elements, which can enhance mutual understanding between driver and back-seat passenger Chao et al.

Furthermore, emoji can also be applied in the area of password security. Kraus et al. Compared with the Standard PIN Personal Identification Number input, a password containing emoji is easier to remember and, thus, emoji-based authentication is a practical alternative to traditional PIN authentication. In the field of communication, research on emoji mainly focuses on two aspects: one is emoji's emotional and linguistic functions in CMC, the other is how different factors, such as individual characteristics, cultural background and system platform, influence users' preferences for emoji use.

For example, Jaeger and Ares analyzed the emotional attributes expressed by 33 facial emojis, and found that most emoji contained one or more emotional meanings. Based on their emotional distribution, Petra et al. Similar studies have found that users tend to use more emoji more in positive messages than negative messages. Both facial and non-facial emoji exhibit a great deal of ability when it comes to expressing emotions Herring and Dainas, ; Jaeger et al.

At the same time, different combinations of emojis can enrich the meanings of emotional expression. More research in this area is referred to in section The Emotional Functions of Emoji of this paper. Individual use of emoji is influenced by many factors. The existing research can be divided into three categories: individual characteristics, cultural background and system platform.

First, the use of emoji is strongly influenced by demographic characteristics such as gender and age of users. Women use them more frequently and men use them more abundantly Tossell et al. Women use emoji more in public communication, but less in private communication Li et al.

In terms of social cognition, emoji with stronger emotional meanings are considered more appropriate and lovely for women than for men, while emoji with weaker emotional meanings but friendlier meanings are considered more appropriate for men Derks et al. Secondly, the use of emoji is closely related to the user's cultural background. Users in different countries will use emoji with specific national or ethnic meanings Gaspar et al.

Users in different countries tend to use emoji differently. Chinese people use emoji and emoticons more often than Spaniards Lin and Chen, Emoji show a high degree of contextual sensitivity and different language types influence the use of emoji. For example, the use of emoji displays a strong correlation among English-speaking countries, while displaying lower correlation among other languages such as Italian and Spanish Barbieri et al.

Finally, different system platforms also lead to differences in emoji usage. Although emoji has a Unicode in the operating system platform, users show emoji differently in IOS, Android and Microsoft operating systems due to the limitation of these software's developmental compatibility Cramer et al. Different social networking platforms such as Twitter, Facebook, Gab and Instagram also have their own particular patterns of emoji usage. For example, users tend to use emoji more frequently and positively on Twitter than on Facebook Hall and Pennington, Gab users tend to use positive emoji to express negative emotions, thus showing irony Settanni and Marengo, More research is referred to in section Diversity of Emoji Use of this paper.

Due to their visual and emotional attributes, emoji can be used in marketing activities. Emoji play an important role in attracting attention, stimulating social interactions and enhancing the experience of consumers, along with their willingness to purchase Das et al.

So it is hardly surprising that emoji are frequently used in consumer interactions Lee et al. Furthermore emoji are also used to depict consumer emotions Li et al. Their dominance in emotional expression makes them an effective tool to measure user's emotions. Textual paralanguages like emoticons and emoji, can influence the cognition and behavior of consumers in marketing activities Luangrath et al.

It has been found that using emoji can enhance the explanatory power, attractiveness, creativity and innovation of marketing activity. With the introduction of emoji in online marketing, more young people are attracted Yakin and Eru, Ge and Gretzel indicate that social media influencers people who take on the dual roles of marketer and active user of social media can initiate online interaction by presenting emoji individually or in combination, which can attract consumers to participate in interactions.

Emoji can also be a way of reflecting consumers' emotions, describing user's profiles Moreno-Sandoval et al. It has been found that gender, age and frequency of usage do not affect consumers' ability to describe and distinguish stimuli with emoji Jaeger et al. In addition, emoji and emoticons are considered simple and intuitive ways to express food-related emotions Vidal et al. Marketers use emoji questionnaires as a common tool to measure user's emotions Jaeger et al.

However, some researchers point out that although emoji show more discriminability and simplicity than emotional words in emotional measurement, their multiple meanings could pose a barrier to the survey. Therefore, emoji questionnaires can't directly replace the existing text-based forms of sentiment survey directly. They can, however, act as a complement to the current form Jaeger et al. In the field of behavioral science, research on emoji focuses on three aspects: motivation, preference and influencing factors.

There has been abundant research focusing on the motivations behind emoji usage. This research has found that emoji are used for managing and maintaining interpersonal relationships Chairunnisa and Benedictus, ; Riordan, b ; Albawardi, , expressing oneself Kaye et al. As a contextual cue, emoji can help users establish an emotional tone, reduce the ambiguity of semantic expression and improve appropriateness relative to context Kaye et al.

There are two aspects of emoji usage preference. One is users' selection of emoji content and the other is the degree to which there is a match between emotions expressed by emoji and real sentiments. For example, users in different countries introduce elements which are representative of their countries into emoji Sadiq et al.

In the field of linguistics, research focuses on the pragmatic functions of emoji and the possibility that they could become an independent language. Emoji have been identified as having semantic properties, and can be used both as an independent language and as a component of a paralanguage providing users with a means of communication and promoting speech acts and interaction Jibril and Abdullah, ; Alshenqeeti, ; Na'aman et al.

There are pros and cons regarding whether emoji can become an independent language. An application was developed to verify the possibility of emoji-first communication Khandekar et al. Other researchers think emoji can't be regarded as an independent language because their meaning largely depends on surrounding text, and only when they are combined with the text can complete semantics be expressed Zhou et al. Studies in this field mainly focus on two aspects. One is the relationship between individual psychological characteristics and emoji usage, and the other is the introduction of emoji into the scale design and the implementation of new psychological measurement tools.

Emoji usage was found to be closely related to some psychological traits such as the big five personality traits, self-monitoring, emotional stress, and others Derks et al. For example, research has shown that frequency of emoji use correlates with emotional stability, extroversion and agreeableness in the big five personality traits, but not with conscientiousness and openness Li et al. At the same time, some studies have attempted to introduce emoji into psychometric scales and have achieved good results in actual measurements Marengo et al.

In the field of medicine and public health, studies on emoji mainly focus on correcting personal behavior and improving doctor-patient communication. Emoji can be used to guide people's behavior regarding health, and it has been shown that using emoji can reinforce correct behavior when it comes to hand hygiene monitoring Gaube et al. Furthermore, using emoji can improve communication between doctors and patients and also enhance patients' abilities to manage their own health Balas et al.

Some researchers suggest developing a set of emoji specifically to be used for patient care, which could help patients better understand and communicate the challenges they face in health management Skiba, In addition, emoji can be used for the identification and prediction of mental illness due to their strength in emotional expression. Marengo et al. In the field of education, research focuses on the impact of emoji on learning efficiency. It has been found that the use of emoji in classroom activities will help students better understand what they have learned Brody and Caldwell, , especially in computer-mediated teaching online learning Dunlap et al.

Emoji can help young children understand abstract concepts such as security, interpersonal management and emotions and also improve their ability to express themselves Fane, ; Fane et al. Understanding users' real emotions when they use emoji is important for future research. At present, it is difficult to accurately measure participants' true reactions through self-reporting.

Categorizing emotions by amassing a corpus using big data is unable to depict users' complex emotions such as are expressed by emoji at a more detailed level, for example emotions such as shame, anger and so on. Therefore, we hold the opinion that in the future, researchers can use some psychological methods in the corpus test to measure the physiological indexes of participants with professional equipment such as nuclear magnetic resonance, electroencephalography and multipurpose polygraphs to depict users' real emotions more accurately.

Future research could also benefit from a more qualitative approach, such as interviews and case studies to learn about emoji use in the context of real-world communication. In practice, some researchers suggest that video and screen shots can be used in concrete operations to observe and record users' choices of emoji during communication Gibson et al. We believe that observing whether users' actual facial expressions differ from their selected emoji emotionally in communication can help researchers understand users' psychological mechanism in communication.

At present, research focuses on the description of users' preference for emoji, but fails to go deeply into the underlying reasons. Users' preferences for emoji are influenced by many factors such as contextual information, interpersonal relationships, familiarity with emoji and personal interpretations other than official definitions, which are all worthwhile factors to explore. The emergence and widespread use of stickers has impacted the status of emoji, and some research has begun to improve the user experience of stickers Shi et al.

Whether stickers will replace emoji is an interesting topic for researchers. Under the impact of stickers, how to further enhance emoji's performance in emotion and semantic expression and improve user experience is also a direction worth exploring. As part of popular culture, the development and use of emoji reflects specific political and cultural characteristics.

Many researchers have interpreted emoji's social influence from different perspectives. For example, some uncivilized use of emoji can harm public consciousness, a point which is not yet appreciated by the public Zerkina et al.

Other researchers believe that the popularity of emoji reflects multicultural communication and cultural globalization Skiba, , and that there is some unconscious power behind the use of non-verbal cues like emoji Elder, , which strengthen the inequality and exploitation of our social system Stark and Crawford, For example, Leslie argues that the quantitative use of emoji in the workplace such as the use of emoji to give ratings has turned the employee into something like an on-the-shelf item in a digital economy warehouse, affecting their freedom.

The democratization of emoji selection and Unicode should also be discussed. Emoji of different skin colors have been introduced to address the lack of racial representation Sweeney and Whaley, Therefore, future research can explore the deeper meaning of emoji use from different perspectives, especially the links between emoji use and political movements, subcultural groups, and social inequality.

This paper systematically reviews related research on emoji, aiming to provide a global perspective and clues for researchers interested in emoji. This paper summarizes the developmental process, usage features, functional attributes, and fields of research related to emoji. Emoji developed from emoticons, and have both emotional and semantic functions. The use of emoji is influenced by and varies according to factors such as individual circumstances, culture, and platforms.

Ambiguity and misunderstanding may occur in different situations and cultural backgrounds. From the perspective of many fields communication, computing, behavioral science, marketing, and education , this paper comprehensively combs the research topics, methods and tools used in studies related to emoji, systematically summarizes the research status of emoji in various fields, and puts forward some new perspectives for future emoji research such as emotional association, use preference, new modalities and impacts on society.

The datasets generated for this study are available on request to the corresponding author. QB contributed conception and design of the study. All authors contributed to manuscript revision, read, and approved the submitted version. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Ai, W. Google Scholar. Al Rashdi, F. Functions of emojis in WhatsApp interaction among Omanis.

Discourse Context Media 26, — Al-Azani, S. Albawardi, A. The translingual digital practices of Saudi females on WhatsApp. Discourse Context Media 25, 68— Aldunate, N. An integrated review of emoticons in computer-mediated communication. Alshenqeeti, H. Are emojis creating a new or old visual language for new generations?

A socio-semiotic study. Aoki, S. Archer, D. Words and everything else: verbal and nonverbal cues in social interpretation. Aull, B. A study of phatic emoji use in WhatsApp communication. Internet Pragmat. Ayvaz, S. The effects of emoji in sentiment analysis. Balas, B. Thin-slice perception develops slowly. Child Psychol. Barbieri, F. Revealing patterns of Twitter emoji usage in Barcelona and Madrid. Berengueres, J. Bich-Carriere, L.

Say it with a smiling face with smiling eyes: judicial use and legal challenges with emoji interpretation in Canada. Semiotics Law 32, — Boothe, D. Brody, N. Cues filtered in, cues filtered out, cues cute, and cues grotesque: teaching mediated communication with emoji pictionary. Butterworth, S. Sender gender influences emoji interpretation in text messages. Cahyaningtyas, R. Cappallo, S. New modality: emoji challenges in prediction, anticipation, and retrieval.

IEEE Trans. Multimedia 21, — Carvalho, P. Chairunnisa, S. Analysis of emoji and emoticon usage in interpersonal communication of Blackberry messenger and WhatsApp application user. Chao, C. Chen, X. Exploring user behaviour of emoticon use among Chinese youth.

Chen, Y. Chen, Z. Cheng, L. Do I mean what I say and say what I mean? Fonseca J. Chik, A. A comparative multimodal analysis of restaurant reviews from two geographical contexts. Cho, K. The differences of emoticon use and its effects depending upon problem types and discussion message types in the process of online problem-solving discussions. Cramer, H. Daniel, T. Emojis affect processing fluency on social media.

Media Cult. Das, G. To emoji or not to emoji? Examining the influence of emoji on consumer reactions to advertising. Derks, D. Emoticons and online message interpretation. The role of emotion in computer-mediated communication: a review. Dimson, T. Emojineering part 1: machine learning for emoji trends.

Blog Donato, G. Dresner, E. Functions of the nonverbal in CMC: emoticons and illocutionary force. Theory 20, — Dunlap, J. Elder, A. What words can't say: emoji and other non-verbal elements of technologically-mediated communication. Ethics Soc. Esposito, G. Nudging to prevent the purchase of incompatible digital products online: an experimental study. Fane, J. Using emoji as a tool to support child wellbeing from a strengths-based approach. Contexts 21, 96— Exploring the use of emoji as a visual research method for eliciting young children's voices in childhood research.

Early Child Dev. Care , — Felbo, B. Fernandez-Gavilanes, M. Creating emoji lexica from unsupervised sentiment analysis of their descriptions. Expert Syst. Gallo, K. A focus group approach to understanding food-related emotions with children using words and emojis. Ganster, T. Same same but different!? The differential influence of smilies and emoticons on person perception.

Gantiva, C. Cortical processing of human and emoji faces: an ERP analysis. Gaspar, R. Crisis as seen by the individual: the Norm Deviation Approach. Psyecology 6, — Beyond positive or negative: qualitative sentiment analysis of social media reactions to unexpected stressful events.

Gaube, S. How a smiley protects health: a pilot intervention to improve hand hygiene in hospitals by activating injunctive norms through emoticons. Gawne, L. Emoji as digital gestures. Language Internet Ge, J. Emoji rhetoric: a social media influencer perspective.

Gibson, W. Discourse Context Media 26, 91— Guthier, B. Hall, J. Self-monitoring, honesty, and cue use on Facebook: the relationship with user extraversion and conscientiousness. Hallsmar, F. Multi-class sentiment classification on Twitter using an emoji training heuristic. Data Syst. Harris, R.

An investigation of the computer-mediated communication of emotions. Hayati, S. Herring, S. Hjartstrom, H. Distraction and facilitation: the impact of emotional sounds in an emoji oddball task. PsyCh J. Hogenboom, A. Jack, R. Cultural confusions show that facial expressions are not universal. Jaeger, S. Dominant meanings of facial emoji: insights from Chinese consumers and comparison with meanings from internet resources.

Food Qual. Measurement of product emotions using emoji surveys: case studies with tasted foods and beverages. Measuring consumers' product associations with emoji and emotion word questionnaires: case studies with tasted foods and written stimuli. Food Res. Valence, arousal and sentiment meanings of 33 facial emoji: insights for the use of emoji in consumer research. Can emoji be used as a direct method to measure emotional associations to food names?

Preliminary investigations with consumers in USA and China. Jiang, F. Microblog sentiment analysis with emoticon space model. Jibril, T. Relevance of emoticons in computer-mediated communication contexts: an overview. Asian Soc. Juhasz, A. Mobile phone use in romantic relationships. Marriage Fam. Kaneko, D. EmojiGrid: a 2D pictorial scale for cross-cultural emotion assessment of negatively and positively valenced food.

Kaye, L. Kelly, R. Khandekar, S. Kim, J. Kimura, M. Kraus, L. LeCompte, T. Lee, D. Social Science Electronic Publishing. Lee, J. Lee, V. The effect of social presence on the facial and verbal expression of emotion and the interrelationships among emotion components. Nonverbal Behav. Leslie, E. This other atmosphere: against human resources, emoji and devices. Li, M. Multi-class Twitter sentiment classification with emojis.

Li, W. Li, Y. Enhancing targeted advertising with social context endorsement. Liebeskind, C. Lim, S. On stickers and communicative fluidity in social media. Lima, M. Can front-of-pack nutrition labeling influence children's emotional associations with unhealthy food products?

An experiment using emoji. Lin, T. A preliminary study of the form and status of passionate affection emoticons. Liu, K. Lo, S. The nonverbal communication functions of emoticons in computer-mediated communication. Luangrath, A. Textual paralanguage and its implications for marketing communications. The logic of design research. Learning: Research and Practice , 4 2 , — Horizon report: higher education edition. Feghali, T. A web-based decision support tool for academic advising.

Feng, S. Research on data mining in university admissions decision-making. International Journal of Advancements in Computing Technology , 3 6 , — Fleiss, J. Statistical methods for rates and proportions. New York: Wiley. Garcia-Gorrostieta, J. Automatic argument assessment of final project reports of computer engineering students. Computer Applications in Engineering Education, 26 5 , — Ge, C.

Application of grey forecasting model based on improved residual correction in the cost estimation of university education. International Journal of Emerging Technologies in Learning , 10 8 , 30— Gierl, M. Automated essay scoring and the future of educational assessment in medical education.

Medical Education , 48 10 , — Gough, D. An introduction to systematic reviews , 2nd ed. Gutierrez, G. Mining: Students comments about teacher performance assessment using machine learning algorithms. Hall Jr. Customized content delivery for graduate management education: Application to business statistics.

Journal of Statistics Education , 16 3. Haugeland, J. Artificial intelligence: The very idea. Cambridge, Mass. Hew, K. British Journal of Educational Technology , 50 3 , — Hinojo-Lucena, F. Artificial intelligence in higher education: A bibliometric study on its impact in the scientific literature.

Education Sciences , 9 1 , Hoffait, A. Early detection of university students with potential difficulties. Decision Support Systems , , 1— Hooshyar, D. A flowchart-based intelligent tutoring system for improving problem-solving skills of novice programmers. Journal of Computer Assisted Learning , 31 4 , — Howard, C.

Shifting the load: A peer dialogue agent that encourages its human collaborator to contribute more to problem solving. International Journal of Artificial Intelligence in Education , 27 1 , — Howard, E. Contrasting prediction methods for early warning systems at undergraduate level. Internet and Higher Education , 37 , 66— Huang, C.

Developing an intelligent diagnosis and assessment e-Learning tool for introductory programming. Huang, J. The research and design of web-based intelligent tutoring system. International Journal of Multimedia and Ubiquitous Engineering , 11 6 , — Huang, S. Effects of using artificial intelligence teaching system for environmental education on environmental knowledge and attitude. Hussain, M. Student engagement predictions in an e-Learning system and their impact on student course assessment scores.

Computational Intelligence and Neuroscience. Iglesias, A. Reinforcement learning of pedagogical policies in adaptive and intelligent educational systems. Jackson, M. Is intelligent online tutoring software useful in refreshing financial accounting knowledge? Jain, G. Jeschike, M. Equipping virtual laboratories with intelligent training scenarios. Jia, J. AI Magazine , 30 2 , 59— Jonassen, D.

Constructivism and computer-mediated communication in distance education. American Journal of Distance Education , 9 2 , 7— Kalz, M. A model for new linkages for prior learning assessment. Campus-Wide Information Systems , 25 4 , — Using an e-Learning system with integrated concept maps to improve conceptual understanding. International Journal of Instructional Media , 37 2 , — Kaplan, A. On the interpretations, illustrations, and implications of artificial intelligence.

Business Horizons , 62 1 , 15— Kardan, A. A decision support system for course offering in online higher education institutes. International Journal of Computational Intelligence Systems , 6 5 , — Prediction of student course selection in online higher education institutes using neural network.

Computers and Education , 65 , 1— Kose, U. International Journal of Engineering Education , 32 1, A , — Li, X. Intelligent agent-supported online education. Decision Sciences Journal of Innovative Education , 5 2 , — Lo, J. Computers and Education , 58 1 , — Lodhi, P.

StuA: An intelligent student assistant. Luckin, R. Intelligence unleashed - an argument for AI in education. Ma, H. McNamara, D. A hierarchical classification approach to automated essay scoring. Assessing Writing , 23 , 35— Misiejuk, K. State of the field report on learning analytics. SLATE report —2. Miwa, K. An intelligent tutoring system with variable levels of instructional support for instructing natural deduction. Transactions of the Japanese Society for Artificial Intelligence , 29 1 , — Moher, D.

BMJ , , b Nehm, R. Transforming biology assessment with machine learning: Automated scoring of written evolutionary explanations. Journal of Science Education and Technology , 21 1 , — Neumann, W. Social research methods: Qualitative and quantitative approaches. Boston: Pearson. Ng, S. Design of an agent-based academic information system for effective education management.

Information Technology Journal , 10 9 , — Nguyen, J. A linguistic multi-criteria decision-aiding system to support university career services. Applied Soft Computing Journal , 67 , — Nicholas, D. Peer review: still king in the digital age.

Learned Publishing , 28 1 , 15— Oztekin, A. A hybrid data analytic approach to predict college graduation status and its determinative factors. Industrial Management and Data Systems , 8 , — Ozturk, Z. Sentiment analysis: An application to Anadolu University. Acta Physica Polonica A , 3 , — Palocsay, S.

A study of the effectiveness of web-based homework in teaching undergraduate business statistics. Decision Sciences Journal of Innovative Education , 6 2 , — Paquette, L. Designing a knowledge representation approach for the generation of pedagogical interventions by MTTs. Payne, V. Effect of a limited-enforcement intelligent tutoring system in dermatopathology on student errors, goals and solution paths. Artificial Intelligence in Medicine , 47 3 , — Artificial intelligence in education: Challenges and opportunities for sustainable development.

Perez, S. Identifying productive inquiry in virtual labs using sequence mining. Baker, X. Hu, M. Chapter Google Scholar. Perin, D. Assessing text-based writing of low-skilled college students. International Journal of Artificial Intelligence in Education , 28 1 , 56— Petticrew, M. Systematic reviews in the social sciences: A practical guide.

Malden; Oxford: Blackwell Pub. Phani Krishna, K. Student information system and performance retrieval through dashboard. Popenici, S. Exploring the impact of artificial intelligence on teaching and learning in higher education. Research and Practice in Technology Enhanced Learning. Prinsloo, P. E-Learning and Digital Media , 14 3 , — Quixal, M. How can writing tasks be characterized in a way serving pedagogical goals and automatic analysis needs?

Calico Journal , 33 1 , 19— Raju, D. Exploring student characteristics of retention that lead to graduation in higher education using data mining models. Journal of Educational Computing Research , 56 1 , 23— Ray, R. Teaching college level content and reading comprehension skills simultaneously via an artificially intelligent adaptive computerized instructional system. Reid, J. Managing learner support. Lockwood Ed. London: Routledge. Rovira, S. Data-driven system to predict academic grades and dropout.

PLoS One , 12 2 , 1— Russel, S. Artificial intelligence - a modern approach. New Jersey: Pearson Education. Salmon, G. E-moderating - the key to teaching and learning online , 1st ed. Samarakou, M. An advanced eLearning environment developed for engineering learners. International Journal of Emerging Technologies in Learning , 10 3 , 22— Sanchez, E.

Future Internet , 8 2. Schiaffino, S. Sebastian, J. Changing stigmatizing attitudes to mental health via education and contact with embodied conversational agents. Computers in Human Behavior , 73 , — Selwyn, N. Is technology good for education? Shen, V. Intelligent multiagent tutoring system in artificial intelligence. International Journal of Engineering Education , 27 2 , — Measures of diagnostic accuracy: Basic definitions.

Smith, R. Peer review: a flawed process at the heart of science and journals. Journal of the Royal Society of Medicine , 99 , — Spikol, D. Supervised machine learning in multimodal learning analytics for estimating success in project-based learning. Journal of Computer Assisted Learning , 34 4 , — Sreenivasa Rao, K. Educational data mining for student placement prediction using machine learning algorithms.

Steenbergen-Hu, S. Journal of Educational Psychology , 2 , — Sultana, S. Predicting performance of electrical engineering students using cognitive and non-cognitive features for identification of potential dropouts. International Journal of Electrical Engineering Education , 54 2 , — Tai, D. Effective e-learning recommendation system based on self-organizing maps and association mining. Electronic Library , 26 3 , — Tegmark, M. Life 3. London: Penguin Books.

Teshnizi, S. Acta Informatica Medica, 23 5 , Thatcher, S. The use of artificial intelligence in the learning of flight crew situation awareness in an undergraduate aviation programme. Umarani, S. Speech based question recognition of interactive ubiquitous teaching robot using supervised classifier.

Umer, R. On predicting academic performance with process mining in learning analytics. Journal of Research in Innovative Teaching , 10 2 , — Vlugter, P. Computer Assisted Language Learning , 22 2 , — Walsh, K. Human machine learning symbiosis. Welham, D. AI in training — : Foundation for the future or misplaced optimism?

British Journal of Educational Technology , 39 2 , — Weston-Sementelli, J. Comprehension and writing strategy training improves performance on content-specific source-based writing tasks. International Journal of Artificial Intelligence in Education , 28 1 , — Wickham, H. R for data science: Import, tidy, transform, visualize, and model data , 1st ed.

Yang, F. Community-organizing agent: An artificial intelligent system for building learning communities among large numbers of learners. Yang, Y. Computers and Education , 53 3 , — Yoo, J. Can online discussion participation predict group project performance? Investigating the roles of linguistic features and participation patterns. Yuanyuan, J. Development of an intelligent teaching system based on 3D technology in the course of digital animation production. International Journal of Emerging Technologies in Learning , 9 9 , 81— Zhu, W.

Engineering Economist , 60 4 , — Download references. Olaf Zawacki-Richter, Victoria I. You can also search for this author in PubMed Google Scholar. In particular, OZR as the leading author, has made a major contribution to the conception and design of the research; the data collection, screening of abstracts and full papers, the analysis, synthesis and interpretation of data; VIM has made a major contribution to the data collection, screening of abstracts and full papers, the analysis, synthesis and interpretation of data; MB has made a major contribution to the data collection, screening of full papers, the analysis, synthesis and interpretation of data; as a native speaker of English she was also responsible for language editing; FG has made a major contribution to the data collection, and the screening of abstracts and full papers.

Victoria I. Correspondence to Olaf Zawacki-Richter. Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Reprints and Permissions. Zawacki-Richter, O. Systematic review of research on artificial intelligence applications in higher education — where are the educators?. Download citation. Received : 26 July Accepted : 01 October Published : 28 October Skip to main content.

Search all SpringerOpen articles Search. Download PDF. Review article Open Access Published: 28 October Systematic review of research on artificial intelligence applications in higher education — where are the educators?

Abstract According to various international reports, Artificial Intelligence in Education AIEd is one of the currently emerging fields in educational technology. Introduction Artificial intelligence AI applications in education are on the rise and have received a lot of attention in the last couple of years.

What is the nature and scope of AI applications in the context of higher education? Search strategy The initial search string see Table 1 and criteria see Table 2 for this systematic review included peer-reviewed articles in English, reporting on artificial intelligence within education at any level, and indexed in three international databases; EBSCO Education Source, Web of Science and Scopus covering titles, abstracts, and keywords.

Table 1 Initial search string Full size table. Table 2 Final inclusion and exclusion criteria Full size table. Full size image. Results Journals, authorship patterns and methods Articles per year There was a noticeable increase in the papers published from onwards. Conclusions and implications for further educational research In this paper, we have explored the field of AIEd research in terms of authorship and publication patterns. Notes 1. References Acikkar, M.

Article Google Scholar Adamson, D. Article Google Scholar Agaoglu, M. Article Google Scholar Ahmad, H. Article Google Scholar Alfarsi, G. Google Scholar Aluko, R. Article Google Scholar Aluthman, E. Article Google Scholar Amigud, A. Article Google Scholar Andris, C. Article Google Scholar Aparicio, F.

Article Google Scholar Bakeman, R. Book Google Scholar Baker, R. Article Google Scholar Barker, T. Article Google Scholar Biletska, O. Article Google Scholar Blikstein, P. Article Google Scholar Brunton, J. Google Scholar Calvo, R. Article Google Scholar Camacho, D. Article Google Scholar Casamayor, A. Article Google Scholar Chen, J. Article Google Scholar Chodorow, M. Article Google Scholar Chou, C. Article Google Scholar Cobos, C.

Article Google Scholar Cohen, J. Article Google Scholar Contact North. Google Scholar DeCarlo, P. Article Google Scholar Delen, D. Article Google Scholar Dikli, S. Article Google Scholar Dobre, I. Article Google Scholar Dodigovic, M. Article Google Scholar Duarte, M. Article Google Scholar Duffy, M. Article Google Scholar Duzhin, F.

Google Scholar Feng, S. Article Google Scholar Fleiss, J. Article Google Scholar Gierl, M. Article Google Scholar Gough, D. Google Scholar Gutierrez, G. Google Scholar Hall Jr. Article Google Scholar Hoffait, A. Article Google Scholar Hooshyar, D. Article Google Scholar Howard, C.

Article Google Scholar Howard, E. Article Google Scholar Huang, C. Google Scholar Huang, J. Article Google Scholar Hussain, M. Article Google Scholar Jackson, M. Article Google Scholar Jain, G. Article Google Scholar Jeschike, M. Google Scholar Jia, J. Article Google Scholar Jonassen, D. Article Google Scholar Kalz, M. Google Scholar Kaplan, A. Article Google Scholar Kardan, A. Article Google Scholar Kose, U. Google Scholar Li, X.

Article Google Scholar Lo, J. Article Google Scholar Lodhi, P. Article Google Scholar Luckin, R. Article Google Scholar Misiejuk, K. Google Scholar Miwa, K. Article Google Scholar Moher, D. Article Google Scholar Nehm, R. Article Google Scholar Neumann, W. Google Scholar Ng, S. Article Google Scholar Nguyen, J. Article Google Scholar Nicholas, D. Article Google Scholar Oztekin, A. Article Google Scholar Ozturk, Z. Article Google Scholar Palocsay, S. Article Google Scholar Paquette, L.

Article Google Scholar Payne, V. Google Scholar Perez, S. Chapter Google Scholar Perin, D. Article Google Scholar Petticrew, M. Article Google Scholar Popenici, S. Article Google Scholar Quixal, M. Article Google Scholar Raju, D. Article Google Scholar Ray, R. Article Google Scholar Reid, J. Google Scholar Rovira, S. Article Google Scholar Russel, S.

Google Scholar Salmon, G. Google Scholar Samarakou, M.

NANCY TALBOT SCHOLARSHIP ESSAY