Personalizing the Content of Health Information Technology Courses Using Data Mining Techniques

Personalization of educational content based on students' needs, conditions, and preferences is one of the most important trends in educational data mining. This study aimed to personalize the content of some courses based on the students’ views using a data mining technique.

This descriptive cross-sectional study was conducted on 7th semester undergraduate students (in the last semester of theoretical course) (n=56) studying Health Information Technology in Mashhad, Semnan, and Ahvaz Universities of Medical Sciences, Iran. The participants were selected randomly, and the data were collected using a researcher-made questionnaire. Validity of the questionnaire was verified by the Health Information Technology faculty members. The students were asked to suggest the preferred order for the most effective teaching of Health Information Technology course headings 1, 2, and 3. Moreover, they were asked to eliminate the headings that were not necessary.  Subsequently, genetic-based algorithms for data mining were used to extract the most frequent patterns from the presented sequences.

In total, four frequent patterns were extracted from the collected data. The first group of the students suggested a common sequence for Health Information Technology course headings 1, 2, and 3. The second and third groups suggested a common sequence for parts of the “Health Information Technology 1” course. Eventually, no frequent patterns were extracted from the fourth group. Inappropriate headings, sequence of courses, references, content volume, and the method of presentation (theoretical and practical) were the most important factors in obtaining these results.

The analysis of the results by the experts showed that the proposed algorithm was useful in providing appropriate sequences of the educational content.