Title :
Semantic Search Based on SVO Constructions in Chinese
Author :
Gao, Ming ; Wang, Ji-cheng
Author_Institution :
Sch. of Electron. & Inf. Eng., Tongji Univ., Shanghai, China
Abstract :
In Chinese the SVO (subject-verb-object) construction is often appears in the query in information retrieval. The SVO construction will be cut into several separate key words, if the traditional retrieval algorithm is used, it will degenerate to a normal Boolean search, because the semantic meaning may be different when the order of the key words differs, thus the semantic implied in the SVO construction will not be exists, that is to say result of an text-retrieval algorithm will reflected by the order of the key words. For this question: taking the advantages of the structural features of forward and inverted index, the paper gives the definition of query step, document step, analyses their reflection on the retrieval, and proposed the pre-processing algorithm to form the new retrieval model based on VSM. Subsequently, The ontology and semantic web support was given, the procedure of translating user query into formal query for semantic search was also given in detail.
Keywords :
information retrieval; ontologies (artificial intelligence); search engines; semantic Web; document step; formal query; forward index; information retrieval; inverted index; normal Boolean search; ontology; query step; semantic search; semantic web support; subject-verb-object construction; text-retrieval algorithm; Algorithm design and analysis; Indexing; Information retrieval; Information systems; OWL; Ontologies; Reflection; Semantic Web; Tensile stress; Text analysis;
Conference_Titel :
Web Information Systems and Mining, 2009. WISM 2009. International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-0-7695-3817-4
DOI :
10.1109/WISM.2009.51