Mining Interesting Substructures based on Various Relevance Measures

点击翻译按钮获取中文摘要

In this research, by upgrading hyperclique patterns and conditional contrast patterns in itemset mining into graph domain, we propose correlation and contrast link formation patterns in a dynamic network, respectively. Discovery of sets of link formation patterns having opposite characteristics, i.e., correlation and contrast, can expect to obtain deep understanding of target dataset. Experiments using real world datasets confirm the effectiveness of the proposed framework.