基于FCA的搜索引擎的信息检索模型及其实现

摘要

针对目前搜索引擎因对同义或近义关键词识别困难导致查准率降低的问题,提出了基于形式概念分析(FCA)的智能搜索引擎,为其提出一个基于关键词与FCA相结合的信息检索模型,即利用形式背景(Context)来挖掘隐含在文档中的潜在的概念关系,利用概念格将文档根据概念间的泛化-概化关系进行聚类,以便于用户找到最相关的信息,从而有效地提高了信息查找的效率.试验验证了该模型的可行性,通过实例来演示模型的构造和实现.试验结果表明该模型在概念理解,结果分类方面弥补了目前搜索引擎的缺陷.%Existing search engine returns excessive information due to embarrassment of distinguish synonymous or relevant keywords. The paper put forward an intelligent search engine based on Formal Concept Analysis (FCA). The information retrieval model based on FCA analyzed conceptual relativity between documents by FCA theory. Users browse documents organized on Galois (concept) lattices rather than a hierarchy, so the search result can be adjusted according to user's intent. Test validates its feasibility, and test result indicates that this model can improve actual search engines in concept comprehension.