信息爆炸是当今信息社会的一大特点,如何在海量的信息中有效地找到所需信息因而成为了一个关键问题,语义检索技术是解决这一问题非常有潜力的方法.本文对信息检索中的若干关键问题进行了研究,提出了基于语义处理技术的信息检索模型--SPTIR,该模型主要包括以下关键技术:基于词义消歧的语义查询扩展、基于词汇语义相关性度量的查询优化和基于文档语义相关性的检索结果重排序.最后使用大型测试数据集和多项性能指标对SPTIR模型的检索性能进行了试验评估,实验结果充分验证了SPTIR模型的竞争优势以及该模型采用的各项语义处理技术对提高检索性能所起的积极作用.%We are in an information age that is mainly characterized by information explosion, and how to find moreprecise search results in the ocean of information becomes a key issue. Semantic search technique, fortunately, is a hopeful way to solve this problem. Several key problems in Information Retrieval (IR) domain are addressed and a novel Semantic Processing Technology based Information Retrieval ( SPTIR) model is proposed in this dissertation. SPTIR includes the following key technologies;Word Sense Disambiguation (WSD) based semantic query expansion, word semantic relatedness based query optimization and document semantic relevance based search results re-ranking. Finally large test data sets and a number of performance indicators are used to test the retrieval performance of the proposed model, and the experimental results fully validated the competitive advantage of SPTIR as well as the active role of semantic processing techniques adopted in improving the retrieval performance.
展开▼
