This research aimed to determine the trend towards the use of Chatbots in higher education. 539 documents published in the SCOPUS database were analyzed with Bibliometrix, an open source tool from bibliometric analysis methods, and Biblioshiny, a rich web application. Findings from the research revealed that documents related to the use of Chatbots in higher education were published in the SCOPUS database for the first time in 2015. The most productive year was determined to be 2024. The most productive source in the field is "LECTURE NOTES IN COMPUTER SCIENCE (INCLUDING SUBSERIES LECTURE NOTES IN ARTIFICIAL INTELLIGENCE AND LECTURE NOTES IN BIOINFORMATICS) with 28 documents, and the most relevant author is "Tan, S.". The most productive countries in the field are "USA, INDIA, UK, GERMANY, INDONESIA, CHINA, SPAIN, MALAYSIA, AUSTRALIA and UKRAINE", while the most relevant institution is "UNIVERSITY OF GRANADA". The most cited document on a global scale was determined as "RUDOLPH J, 2023, J APPL LEARN TEACH-a" with DOI number 10.37074/jalt.2023.6.1.9, while the most frequently used words in the documents were determined as "Chatbots" and "higher educations". In higher education, it is recommended that authors who will conduct research on the use of chatbots take into account the findings obtained from this research.
Chatbots are actively used alongside AI technology today. Three AI chatbot prototypes were introduced
at the University of Warwick in 2019, and their history is presented. These bots were developed to
provide comprehensive contributions to postgraduate simulation games, educational application use, and
assistance [
Another study, examining 24 studies published between 2022 and 2023, analyzed higher education
students’ perceptions and usage patterns of AI chatbots, revealing their concerns about accurate
information, critical thinking, and creativity [
An examination of ChatGPT usage and student perceptions using data from 5894 students at Swedish
universities reveals attitudes that difer by gender and field, highlighting the need to develop local,
student-focused AI tools [
The socio-emotional and relational factors of chatbots used in academic advising by university
students in the UAE are examined, and ease of use and social impact are considered important, while
recommendations are ofered on trust and ethics [
The remainder of the paper is structured as follows: Section 2 presents the methodology, Section 3 reports the findings, and Section 4 provides the discussion and conclusions.
To achieve the research purpose, the following questions were answered: 1. What is the distribution of publications in terms of main information, such as year, document type, etc.? 2. What is the distribution of studies by year? 3. What are the most relevant sources? 4. Who are the most relevant authors? 5. Which countries are the most productive? 6. Which institutions are the most relevant? 7. What are the most cited documents and the most frequently used words in these documents?
The purpose of this study is to determine the trend towards the use of Chatbots in higher education
and to guide future studies. In the study where bibliometric analysis methods were preferred, the
SCOPUS database was preferred for the data. SCOPUS; It was chosen because it analyzes a reliable
and rich pool of information that provides independent data and measurements on research areas,
authors, and institutions1. To achieve the purpose of the study, 539 documents were obtained by
scanning with the keywords "Chatbots" AND "Higher Education". The search was carried out in the
SCOPUS database by selecting the "Article title, Abstract, Keywords" search option and ended on July 4,
2025. To perform bibliometric analysis, bibliometrix, an open source tool developed by Massimo Aria
and Corrado Cuccurullo [
General features of the published documents regarding the use of Chatbots in higher education are given in Table 1. 1Scopus https://www.scopus.com/home.uri 2Bibliometrix https://www.bibliometrix.org/home/index.php/layout/bibliometrix
Documentation on the use of Chatbots in higher education began to be published for the first time in 2015. The total number of documents published until July 4, 2025, is 539 and was included in a total of 332 sources. It was determined that 1707 authors carried out studies on the use of Chatbots in higher education, and 61 authors were authors of single-authored documents. From the findings obtained, it is seen that the authors mostly prefer to carry out collaborative studies. When the type of documents published in the SCOPUS database was examined, it was determined that the authors mostly published in the "article" document type ( = 241). Other published document types were determined as "conference paper" ( = 194), "book chapter" ( = 58), "conference review" ( = 22), "review" ( = 11), "book" and "editorial" ( = 5), respectively. The least preferred document type was determined to be "erratum" ( = 3).
The distribution of documents published in the SCOPUS database by years is given in Table 2.
As seen in Table 2, the first documents on the use of Chatbots in higher education began to be published in the SCOPUS database in 2015 ( = 1). No documents were found in 2016 and 2017. 5 documents were published in 2018, 10 in 2019, and 11 in 2020. It is seen that the documents started to increase quantitatively as of 2021 ( = 28). The most productive year was determined as 2024 ( = 215). While it was determined that 137 documents related to the field were published in 2025, it is thought that this number will increase by the end of the year. In light of the findings, it can be said that studies on the use of Chatbots in higher education are becoming increasingly important and production will increase further in the coming years.
The 10 most relevant sources in the SCOPUS database, which publish documents on the use of Chatbots in higher education, are given in Table 3.
When the top 10 sources that published the most on the field between 2015 and July 2025 were examined, it was determined that the most published documents were 28. "Education And Information Technologies" Chatbots in higher education While “Lecture Notes In Networks And Systems” published 16 documents for its use, the other sources most relevant to the field are “ACM International Conference Proceeding Series” ( = 11), “Computers And Education: Artificial Intelligence” ( = 11), “IEEE Global Engineering Education Conference, EDUCON”. ( = 11), “Education Sciences” ( = 10), “Journal Of Applied Learning and Teaching” ( = 9), “Communications In Computer And Information Science” ( = 5), and “Frontiers In Education” ( = 5).
The findings will shed light on researchers who will study the use of Chatbots in higher education in terms of accessing the most relevant resources.
The authors who have conducted research on the use of Chatbots in higher education and are most relevant to the field are given in Table 4.
As seen in Table 4, the author most relevant to the field is "Tan, S." with 8 articles. The other authors most relevant to the field are; “Rudolph, J.” ( = 6), “Chen, Y.” ( = 5), “Abbas, N.” ( = 4), “Al Yakin, A.” ( = 4), “Gupta, S.” ( = 4), “Tsivitanidou, O.” ( = 4), “Aeni, N.” ( = 3), “Barranco, F.” ( = 3) and “Bell, D.” ( = 3). It is recommended that authors who will work in this field in the future take the documents of the researchers whose names are listed in Table 4 as examples.
This study also aimed to determine the countries that focus on studies on the use of Chatbots in higher education. As a result of the analysis, the top 10 most productive countries in the field are given in Table 5.
As seen in Table 5, the most productive country in the field was determined to be the USA with 241 publications. The other most productive countries are: INDIA ( = 140), UK ( = 89), GERMANY ( = 87), INDONESIA ( = 85), CHINA ( = 70), SPAIN ( = 68), MALAYSIA ( = 56), AUSTRALIA ( = 46), and UKRAINE ( = 46). It is recommended that the policies and practices of the top 10 most productive countries in the field be taken as an example by other countries.
As seen in Figure 1, the productivity years of the countries are distributed between 2015 and 2023. The most productive year has been determined as 2023 or later. In this context, it can be said that countries that carry out studies on the use of Chatbots in higher education will show real productivity after 2023. The list of the top 10 countries with the most citations is given in Table 6.
In the study, it was determined that the top 10 countries with the most citations were the USA, UK, CHINA, AUSTRALIA, UNITED ARAB EMIRATES, SAUDI ARABIA, GERMANY, QATAR, MALAYSIA, and HONG KONG, respectively.
The research aimed to determine the most relevant institutions for the use of Chatbots in higher education. The list of the top 10 most relevant institutions is given in Table 7.
The findings revealed that the most relevant institution in the field was the "UNIVERSITY OF GRANADA" with 20 articles. The other most relevant institutions were determined as "EAST WEST UNIVERSITY" ( = 17), "THE UNIVERSITY OF JORDAN" ( = 17), "UNIVERSITY OF LEEDS" Afiliation UNIVERSITY OF GRANADA EAST WEST UNIVERSITY THE UNIVERSITY OF JORDAN UNIVERSITY OF LEEDS FERGANA STATE UNIVERSITY BINA NUSANTARA UNIVERSITY TECHNICAL UNIVERSITY OF DENMARK UNIVERSITY OF OULU SOEGIJAPRANATA CATHOLIC UNIVERSITY THE UNIVERSITY OF HONG KONG ( = 13), "FERGANA STATE UNIVERSITY" ( = 11), "BINA NUSANTARA UNIVERSITY" ( = 10), "TECHNICAL UNIVERSITY OF DENMARK" ( = 10), "UNIVERSITY OF OULU" ( = 10), "SOEGIJAPRANATA CATHOLIC UNIVERSITY" ( = 9), and "THE UNIVERSITY OF HONG KONG" ( = 9). It is recommended that other institutions conducting research in this field continue to collaborate with institutions listed as "most relevant institutions" and those that have produced the most documents.
The top 10 globally cited documents related to the use of Chatbots in higher education are given in Table 8.
As seen in Table 8, the most cited document globally was "RUDOLPH J, 2023, J APPL LEARN TEACHa" (Total Citations = 982) with DOI number 10.37074/jalt.2023.6.1.9. The other documents in the top 10 and most cited globally are; “RUDOLPH J, 2023, J APPL LEARN TEACH, DOI: 10.37074/jalt.2023.6.1.23” (Total Citations= 438), “KING MR, 2023, CELL MOL BIOENG, DOI: 10.1007/s12195-022-00754-8” (Total Citations= 405), “MICHEL-VILLARREAL R, 2023, EDUC SCI, DOI: 10.3390/educsci13090856” (Total Citations= 354), “CHEN Y, 2023, INF SYST FRONT, DOI: 10.1007/s10796-022-10291-4” (Total Citations= 307), “ABULIBDEH A, 2024, J CLEAN PROD, DOI: 10.1016/j.jclepro.2023.140527” (Total Citations= 271), “DEMPERE J, 2023, FRONT EDUC, DOI: 10.3389/feduc.2023.1206936” (Total Citations= 201), “WANG T, 2023, APPL SCI, DOI: 10.3390/app13116716” (Total Citations= 199), “WU R, 2024, BR J EDUC TECHNOL, DOI: 10.1111/bjet.13334” (Total Citations= 191), and “IMRAN M, 2023, CONTEMP EDU TECH, DOI: 10.30935/cedtech/13605” (Total Citations= 184). These documents, which are cited worldwide, are thought to make a great contribution to the field. It is recommended that researchers take these documents into consideration and benefit from them for similar studies.
When the most frequently used words in the documents were examined, it was determined that the most preferred word by the researchers was Chatbots (Occurrences = 218). The other words most frequently used in the documents are; “high educations” (Occurrences=162), “students” (Occurrences=157), “artificial intelligence” (Occurrences=75), “engineering education” (Occurrences=54), “chatgpt” and “teaching” (Occurrences=46), “contrastive learning” (Occurrences=38), “adversarial machine learning” and “curricula” It was determined as (Occurrences=36) (See Table 9).
It is recommended that researchers use these words for their literature review in the field.
An examination of the study findings reveals that the use of AI-based chatbots, particularly in higher
education, is increasing, and scientific productivity in this field is expected to reach its peak in 2024. The
ifrst publication of chatbot research in the SCOPUS database in 2015 demonstrates that this technology
represents a new field of application in education. Early prototypes developed at the University of
Warwick [
Following the publication of ChatGPT, a significant contribution to the literature has been made
in both technical and educational fields [
Generally, when the studies are examined, it is seen that chatbots support personalized learning,
consulting, and evaluation processes, but their negative aspects, especially on providing accurate
information, ethical concerns, and critical thinking skills, are also discussed [
When all the findings are examined, the most productive year was determined to be 2024. The most productive source in the field is "LECTURE NOTES IN COMPUTER SCIENCE (INCLUDING SUBSERIES LECTURE NOTES IN ARTIFICIAL INTELLIGENCE AND LECTURE NOTES IN BIOINFORMATICS) with 28 documents, and the most relevant author is "Tan, S.". The most productive countries in the field are "USA, INDIA, UK, GERMANY, INDONESIA, CHINA, SPAIN, MALAYSIA, AUSTRALIA and UKRAINE", while the most relevant institution is "UNIVERSITY OF GRANADA". The most cited document on a global scale was determined as "RUDOLPH J, 2023, J APPL LEARN TEACH-a" with DOI number 10.37074/jalt.2023.6.1.9, while the most frequently used words in the documents were determined as "Chatbots" and "higher education." It is recommended that authors researching the use of chatbots in higher education take into account the findings obtained from this research. In this context, it is important that new studies follow both technical developments and how they can be used ethically by addressing student needs. This study aims to guide students in developing strategies for the ethical use of chatbots in higher education by presenting existing studies and new research trends. Furthermore, considering the use of Bibliometrix, future studies could be enhanced by incorporating broader databases such as Web of Science (WoS) to ensure more comprehensive coverage of the literature.