A Novel Entity Type Filtering Model for Related Entity Finding
Junsan Zhang, Youli Qu, Shengfeng Tian · International Journal of Computers Communications & Control · 2014
Entity is an important information carrier in Web pages. Searchers often want a ranked list of relevant entities directly rather a list of documents. So the research of related entity finding (REF) is a meaningful work. In this paper we investigate the most important task of REF: Entity Ranking. To address the issue of wrong entity type in entity ranking: some retrieved entities don't belong to the target entity type. We propose a novel entity type filtering model in which the target types are composed of the originally assigned type and the new type which is automatically acquired from the topic's narrative to filter wrong-type entities. For the query, we propose a method to process the original narrative to acquire a new query which is composed of noun and verb phrases. The results of experiments show our novel type filtering model gets a better result than the traditional filtering model at whatever precision and recall. Also the experiment shows the method that we acquire a new query is feasible.