Department of Software Engineering
Permanent URI for this collection
Browse
Browsing Department of Software Engineering by Author "Adabashi, Afaf"
Now showing 1 - 1 of 1
Results Per Page
Sort Options
Item DEVELOPMENT OF AN INTELLIGENT TUTORING SYSTEM USING BAYESIAN NETWORKS AND FUZZY LOGIC(2022-01-24) Adabashi, Afaf; Eryılmaz, Meltem; Yazıcı, AliRecently, there has been a rapid growth in web-based Intelligent Tutoring Systems (ITSs) to support the teaching process with the aim of helping students adaptively navigate through online learning materials. Students who use these systems come from different backgrounds with different needs, preferences and characteristics. Therefore, the challenges for ITSs are their ability to provide dynamic adaptation to each individual user and a user-friendly interface in order to deliver knowledge effectively. The efficiency of ITSs depends on the methods used to collect and examine information related to the characteristics of students and their needs. Moreover, depends on the way in which this information is processed to form an adaptive educational context. There are various artificial intelligence methods such as fuzzy logic and the Bayesian network that facilitate the learning process in order to adapt the course content to meet the goal of each student and which deal with uncertainty in the student assessment process. In this thesis, an intelligent tutoring system, called FB-ITS is proposed using a hybrid method based on fuzzy logic and Bayesian networks techniques to adaptively support students in learning in which the adaptation is achieved by modelling the students according to their knowledge level. FB-ITS takes the advantages of fuzzy logic and the Bayesian network, where the fuzzy logic is used to determine student performance in a particular topic of domain according to her/his prior and current knowledge and the Bayesian network is used to identify the state of the related topics based on the evidence that comes from the fuzzy logic system. The effectiveness of FB-ITS was evaluated by comparing it with the two other versions of ITS that were developed and implemented using fuzzy logic and the Bayesian network separately in addition to it having been evaluated by comparing it with an existing traditional e-learning system. The study was conducted with undergraduate students at Atilim University, Turkey. Three dependent variables were utilized to evaluate the effectiveness of the proposed system, including students’ academic performance, students’ satisfaction, and system usability. The results showed that students who studied using FB-ITS had significantly higher academic performance (82.95) on average compared to other students who studied with ITS using the Bayesian network (79.09), ITS using fuzzy logic (69.77) and the traditional e-learning system (64.33). Regarding the time taken to perform the post-test, the results indicated that students who used the FB-ITS needed less time (7.87 minutes) on average compared to students who used the traditional e-learning system (13.86 minutes). From the results, it could be concluded that the new system contributed in terms of the speed of performing the final exam and high academic success. Additionally, the evaluation of the system showed moderate results in terms of students’ satisfaction and the system usability.