STUDENT MODELING MENGGUNAKAN BAYESIAN NETWORKUNTUK KARAKTERISTIK PEMBELAJARAN (STUDI KASUS: JTETI UNIVERSITAS GADJAH MADA)

Students have different characteristics of learning, in terms of knowledge, interest, learning style, and background. General learning materials for each student might not optimally acceptable. Applying these general learning materials yields an ineffective learning process. Thus, the implementation of learning model that is suitable in term of student characteristic is needed. In this thesis, a student modeling based on Bayesian network is presented. Bayesian network is a graphical modeling tool based on causality in the set of random variables that can be used in variety of applications, such as in the context of education. Bayesian network was implemented by using K2 algorithm to determine the learning style of each student in English course held in JTETI UGM. This research was building a student modeling of the learning style using a Bayesian network method. In this research, the evaluation results of accuracy (classification correctly) reach a value of 62.5%. While, the percentage of values that are classified incorrectly was 37.5%.Thus, the learning styles declared dataset has a fairly high accuracy using Bayesian network