Development of an Algorithm for Intelligent Analysis of Behavioral Classifiers for Personalizing Training

摘要

The use of e-learning and its personalization has recently become a popular trend in education. The purpose of this research is to develop a new data mining algorithm to optimize the performance of a personalized learning system. The input parameters are the behavioral characteristics of students obtained during preliminary testing. By means of the obtained data set and the proposed data analysis algorithm, behavioral classifiers are derived, with the help of which students, previously, can be attributed to a group of excellent students (A) or to a group of students who need to seriously improve their knowledge (D). The proposed algorithm was Tested on undergraduates of the Department of computer science. The developed algorithm for training students showed the productivity of using the proposed approach in the educational process. The results obtained can be used in the construction and planning of an educational program. Using the developed data analysis algorithm, the authors predict the final grades of students and students who are at risk of failing the exam. Further development of the approach proposed in this paper involves taking into account the moral factor and developing a set of incentives for students.