NMiner: A System for Finding Related Entities by Mining a Bimodal Network

摘要: Motivated from related entity finding problem, in this paper, we introduce a novel approach to query answering called “NMiner.” NMiner takes advantage of heuristics find answers complex semantic queries. It uses combination natural language processing techniques parse sentences and extract entities, hypertext structure the documents derive relational information, web data relevant entities as search result candidates. Further, bimodal network is created Content Centric Ranking (CCR) Cumulative Structural Similarity (CSS), are proposed rank candidate entities. Our empirical study on ClueWeb09 corpus (with approximately 25 terabytes documents) shows that both CSS CCR outperform PageRank HITS. Moreover, proved be significant solving problem queries performed against largely unstructured text documents.