A Novel Method for Mining Relationships of entities on Web

Abstract

In this paper, we address an important and difficult problem: mining relationships of entities on web. We propose a novel and efficient framework to solve the aforementioned problem. Different from all the existing methods of mining relationships of entities, our method makes full use of the events that have been extracted already to mine relationships between entities. Pairs (E, T) are employed to represent the close relationships. In addition, our method could mine the deep semantic meaning through analyzing the events that the relationships occur in. In the future, multidimensional search and deep analysis are to be performed on these pairs (E, T) and events.