作者: David Dominguez-Sal , Joan Guisado-Gámez , Josep Lluís Larriba-Pey
DOI:
关键词: Graph (abstract data type) 、 Computer science 、 Natural language 、 Query expansion 、 Natural language processing 、 Artificial intelligence 、 Search engine 、 Information retrieval 、 Ambiguity 、 Knowledge base 、 Vocabulary mismatch
摘要: Keyword based search engines have problems with term ambiguity and vocabulary mismatch. In this paper, we propose a query expansion technique that enriches queries expressed as keywords short natural language descriptions. We present new massive strategy using knowledge base by identifying the concepts, adding relevant synonyms semantically related terms. two approaches: (i) lexical locates concepts in base; and, (ii) topological analyzes network of relations among suggests terms path community analysis graph. perform our expansions versions Wikipedia base, concluding combination both provides improvements system's precision up to more than 27%.