31 January 2025 in Journal of Educational Technology and Online Learning
This study addresses the applicability of generative artificial intelligence (GenAI) within the administrative learner support services at Anadolu University Open Education System, a giga university with more than one million learners in Türkiye. The study reveals the performance differences between a rule-based chatbot and different GenAI-based chatbots using a qualitative case study approach. Learner inquiries that the rule-based chatbot could not answer, and frequently asked questions (FAQs) were posed to ChatGPT and Bing Copilot applications in four different cases. The 708 answers in these scenarios were evaluated by three experts. It was observed that GenAI matched learner questions more effectively than the rule-based chatbot. Additionally, Bing Copilot was more successful in generating responses to learners’ questions from the internet compared to ChatGPT, which utilized the FAQ dataset. The study reveals that Gen-AI based chatbots that can work together with rule-based chatbots are more successful in generating answers. Findings demonstrate the potential of GenAI to provide learners with continuous, more natural, and personalized support. These findings offer significant insights into how administrative support services in mass-scale educational institutions can be transformed.
6 May 2025 in Journal of e-learning Research
This Research Examines The Challenges Learners Face During Course Selection In Open And Distance Learning Frameworks And Evaluates The Effectiveness Of AI-Powered Virtual Academic Advisers In Addressing These Issues. While Distance Education Offers Flexibility, The Absence Of Personalized Guidance Often Complicates Decision-Making. This Study Aims To Develop A Virtual Academic Adviser Aligned With Anadolu University's Registration And Course Selection Regulations And Analyze Its Responses To Learner Inquiries. Using Openai's Mygpt Application, Virtual Advisers Were Trained To Interact With Learners Based On Their Academic Performance. A Qualitative Holistic Single-Case Design Was Applied To Compare Responses From Virtual Advisers With Those Of Chatgpt 3.5 And Chatgpt 4. Findings Suggest That Integrating Virtual Advisers, Accounting For Academic Risks Like Repeated Course Performance, Can Enhance Learner Experiences. The Study Underscores The Potential Of AI-Based Virtual Advisers To Improve Academic Performance And Decision-Making In Course Selection, Advocating Further Research Into AI’s Role In Distance Education.
14 September 2024 in Osmangazi Journal of Educational Research
Anadolu University's Open Education System (OES) accommodates over one million students and has incorporated an AI-based Virtual Assistant for non-academic support since 2022. While OES offers abundant information about its programs on its website, there is a notable absence of support services providing job recommendations related to students' chosen programs. This gap in student support extends to the post-graduation phase, with the Virtual Assistant lacking a concept for guiding students in finding employment opportunities. Recognizing the need for comprehensive assistance, this study sought to leverage AI capabilities to offer job recommendations by extracting information from the objectives of 63 OES programs. The initial inquiry involved requesting AI-generated job recommendations based on the stated objectives of these programs. Subsequently, the Virtual Assistant was tasked with providing insights into the occupation fields associated with OES programs. Analysis of the AI's responses, along with the classification of occupations according to the International Standards of Classifications of Occupations (ISCO) and the International Standard Classification of Education (ISCED), forms the core of this study. Contrary to trends observed in most European countries, the predominant number of graduates in Türkiye emerges from business and management fields. However, the correlation between graduation rates and subsequent job placements appears suboptimal within the labor force and employment landscape. The study advocates for the integration of AI in offering job recommendations, incorporating graduation and employment rates. This approach enables students to seek guidance on suitable programs aligned with their skills, fostering a more informed decision-making process. The study underscores the potential for higher education institutes to share employment and labor force data.
6 May 2026 in The International Review of Research in Open and Distributed Learning
This exploratory case study provides an in-depth analysis of the potential of generative artificial intelligence (GenAI) to enhance academic support in open and distance learning (ODL) systems. The study examined learner experiences with a GenAI-based academic support application in an online web publishing course over a semester, focusing on two phases: free use and structured use. Data were collected through semi-structured interviews and dialogue transcripts from 10 distance learners. Findings highlighted both continuity and transformation in learner practices. In both phases, GenAI was valued for time-saving and accurate responses aligned with course materials. Structured tasks in phase 2 encouraged more purposeful engagement, including systematic self-assessment and information verification. Despite technical challenges such as device incompatibility and occasional hallucinations, learners expressed motivation, satisfaction, and a demand for institutional integration. The results, while preliminary, suggest that GenAI-based academic support holds strong potential for broader implementation in large-scale open universities, offering a pathway to balancing quality, access, and cost in addressing the enduring challenges of mass higher education.
27 January 2024 in Açıköğretim Uygulamaları ve Araştırmaları Dergisi
Anadolu Üniversitesi Açıköğretim Sisteminde (AÖS) ortak ders olarak birden fazla programda yer alan dersler mevcuttur. Yabancı Dil, Türk Dili vb. olan bu dersler, temel olarak bir programa ait olup başka programlarda da okutulmaktadır. Bu derslerin hangi program bünyesinde yürütülmesi gerektiği ile ilgili 2021-2022 öğretim yılından itibaren belirli kriterlere göre bir programa atanmasına yönelik aidiyet çalışması gerçekleştirilmiştir. Araştırmada, programların ders aidiyetlerine göre incelenmesi ve bu aidiyetlerin öğrenenlere kayıt öncesi destek vermesi amaçlanmış olup bu amaç doğrultusunda öğrenenler için bir kayıt destek sisteminin geliştirilmesi planlanmıştır. Kayıt destek sisteminde bir programın ders ağırlığının nasıl olduğu, öğrenenin ilgisini çeken derslerin hangi programlarda okutulduğu gibi bilgilerin sunulmasına yönelik çalışma gerçekleştirilmiştir. Araştırma, nitel araştırma desenlerinden durum çalışması ile desenlenmiş ve veriler, doküman analizi kullanılarak incelenmiştir. Araştırmanın verileri, Anadolu Üniversitesi AÖS’te ders aidiyeti uygulamasının yapıldığı 65 programdaki dersler ve bu derslerin aidiyetleri incelenerek elde edilmiştir. Elde edilen bulgulara göre toplam 1.133 dersten 252’si birden fazla programda okutulurken 881 ders ise sadece bir programda okutulmaktadır. 43 ön lisans programında ders listesindeki derslerin aidiyet benzerlik oranı %48,64 iken lisans programlarında bu oran %56,25’tir. 65 programa göre bu oran %52,17’dir. Yaklaşık olarak her iki dersten birinin ders aidiyetinin başka programda olduğu ve seçmeli ders uygulaması bulunmadığı göz önünde bulundurulduğunda sunulacak kayıt destek sisteminin öğrenenlerin kayıt süreçlerini önemli ölçüde kolaylaştıracağı öngörülmektedir.
21 November 2024 in International Conference on Education Research
At the University of Bologna, one of the pioneers of higher education in Europe and the institution that inspired the name of the Bologna Process, courses are labelled according to the International Standard Classification of Education (ISCED), a statistical classification of vocational fields. In Open and Distance Learning (ODL), where the number of programmes is high in parallel with the number of learners, determining the fields of courses taught is crucial not only for measurement and evaluation processes but also for a detailed examination of statistical information in processes such as enrolment and graduation. Processes such as data classification according to specific categories can be rapidly carried out with the help of artificial intelligence (AI) and can be utilised in administrative processes. This study investigated whether ChatGPT-4, one of the AI applications, could classify 1135 courses taught at Anadolu University's Open Education System (AUOES), which is part of the Bologna Process, according to ISCED fields, considering the content of the courses. In this study, document analysis was applied to the data analysis. According to the results, the highest number of courses in AUOES were in business, administration, and law (386), while the fewest courses were in education (27). These results indicate that courses related to white-collar professions are taught frequently and are influenced by the programmes at AUOES. This study suggests that AI can be used in administrative processes and to classify courses according to ISCED fields. Categorising all courses according to ISCED or a similar standard could enable the analysis of courses in vocational fields. Determining the fields of courses according to certain standardisations in ODL could allow the courses, and consequently, the books and materials, to be handled by subject matter experts. Decision-makers in ODL could plan the teaching of courses in line with needs by considering the employment situation in vocational fields when launching new programmes or updating course lists. Researchers could investigate the accuracy of AI's processes in administrative tasks and gather the opinions of subject matter experts, opening up new avenues for further research and exploration in the field of AI in education.