Data mining techniques for detecting behavioural patterns of gifted students in online learning environment
Abstract
Presented article deals with selected results of data mining analysis of psychological and educational data, collected within 5 years (2006-2011) on the group of 91 children, who participated in a specific online program for gifted children, organised by faculty of Mathematics and Physics, Charles University in Prague, Czech Republic. The information record for each case (student) is represented by 151 variables, of both categorical and metric nature, describing 1) personal characteristics, including motivation and intelligence 2) behavioural and action records, including individual decision making records, and 3) particular educational results and learning path.
The study is based on comparison of students with very similar personal characteristics like motivation and intelligence and their study results. The success in a selected online course seems to be related to the nature of particular student talent.
Student, who had chosen the right courses, that matched his/her nature of talent, succeed. An individual, who for some reason chooses the course that does not match his /her nature of talent, is more likely to fail.