Summarizing Semantic Associations Based on Focused Association Graph

摘要

As the explosive growth of online linked data, there is an urgent need for an efficient approach to discovering and understanding various semantic associations. Research has been done on discovering semantic associations as link paths in linked data. However, few discussions have been given on how we can understand complex and large-scale semantic associations. Generating human understandable summaries for semantic associations is a good choice. In this paper, we first give a novel definition of semantic association, and then we describe how we discover semantic associations by mining link patterns. Next, a notion of Focused Association Graph is proposed to characterize merged associations among a set of focused objects. Then we focus on summarizing of Focused Association Graph. Concise summaries are generated with the help of Steiner Tree problem. Experiments show that our approach is feasible and efficient in generating summaries for semantic associations.