A data mining approach to analyzing student-peer relationships from communication history records

초록

In recent years, bullied students and delinquent groups in teenagers cause many serious social problems. In this paper we propose a novel approach that analyzes peer relationships among students more objectively. As the data for objective analysis, we use communication history records that are collected from various communication tools such as telephones, e-mails, short messages, and messengers. We use a simple intuition that communication history records implicitly contain peer relationship information, and we adopt data mining techniques for the more systematic analysis. Our key contribution is to propose a novel data mining-based approach to identifying the potentially bullied students and potentially delinquent groups based on the objective data of communication history records. The proposed method consists of the following steps. First, we formally define the notion of degree of familiarity between students and present mathematical formulas that compute the degree based on communication history records. We here use the intuition that the degree of familiarity from student x to student y becomes higher as the number of communications from x to y increases. Second, using the degree of familiarity we find out the students who are potentially bullied. This procedure is based on the assumption that a bullied student may have a very small number of communication history records issued from other students. Third, we adopt a clustering technique, one of the most representative data mining techniques, to identify meaningful student groups. To use the clustering technique, we first define the measure of similarity between friends based on the degree of familiarity, and we then perform clustering using the similarity measure. Last, to show the practical use of the proposed method, we have implemented the method and interpreted the meaning of experimental results. Overall, we believe that our research result provides a useful framework that analyzes peer relationships among students more objectively and more systematically. © 2013 ISSN 1349-4198.

키워드