Semantic Information Retrieval Research Based on Co-occurrence Analysis

Qiu Junping,Lou Wen

DOI: https://doi.org/10.3969/j.issn.1001-8867.2012.06.010

2012-01-01

Abstract:In order to improve user experience of information retrieval system,researchers always explore into query optimization algorithms,visualization and other aspects.This paper used literature review,co-occurrence analysis,ontology building and other methods to design a model and process of semantic information retrieval based on co-occurrence analysis.And the paper selected the world's archaeological category data from Wuhan University Library's bibliographic retrieval systems to experimental analysis.Literature review summarized that semantic information retrieval research mainly concentrates in the ontology-based query techniques,semantic annotation and semantic relation retrieval.And most recent system can only realize obvious relations retrieval.This paper constructed a model to realize potential relations retrieval.The model is divided into three parts:Normalizer,Analyzer and Ontology builder.Analyzer is the model core.Experimental analysis proved the feasibility of co-occurrence analysis used in semantic information retrieval.Compared with traditional retrieval,semantic information retrieval based on co-occurrence analysis is more user-friendly and instructive.