Computing the extent of diversity for a frequent pattern
- Student Name: M. Kumara Swamy
- Course: PhD in Computer Science and Engineering
"We have conceptualized the problem of computing the extent of diversity for a frequent pattern. Given a domain, set of items can be grouped into a category and a pattern may contain the items which belong to multiple categories. The existing frequent pattern mining approaches do not distinguish the patterns by analyzing the category of items in the pattern. The notion of diversity captures the extent of items in the pattern belong to multiple categories. We propose a framework to rank the pattern by analyzing the extent the items of the pattern belong to multiple categories in the corresponding concept hierarchy. We defined the notion of diversity to rank the patterns called diverse rank (drank) to capture the extent of diversity. We propose two frameworks to compute the diversity of the pattern: one is by considering balanced concept hierarchy of items as an input and the other is by considering unbalanced concept hierarchy of items as an input. We provide applications by exploiting the notion of diversity and show that such knowledge can be exploited to build efficient recommendation system and web search systems. "