An Ontology-Based Approach To Storage And Represent The Results Of Text Mining

ObjectiveWith the development of functional genomics, proteomics, metabonomics,the number of biomedical literatures is increasing rapidly. So,how to access to the needed information from such a large pool of information fastly, accuratly, and conveniently is an important issue facing information users. The appearance of text mining makes us promising and brilliant.How can we reasonably represent and effectivly organize the outcomes of the text mining,such as semantic information ,thus facilitating the biomedical information uses? After the Semantic Web is proposed,the advantages of ontology in knowledge representation and knowledge organization become more prominent. The reseach of ontology in domain knowledge represention and management become a topic. The main purpose of this paper is to try to represent the semantic information obtained by text mining with ontology.Materials and methodsThe category subject heading "diseases category" with all allowable qualifiers was chosen when searching MEDLINE .The other two categories were also like this. Then high frequency major MeSH terms were calculated among related articles for co-word clustered analysis, and MeSH pairs were extracted as rules under description and evaluation. Articles that containing the pair of MeSH terms were read and the relationship of the subject headings was judged. Semantic relations in Universal Medical Language System (UMLS) were adapted to link two MeSH terms that formed a association rule. In order to limit quantity of rules, medical subjects headings were substituted with subclass number of MeSH tree structure. Domain ontology about neoplasm was built under protege with the semantic information obtained by text mining.ResultsAt last 346 association rules were extracted: 202 about anatomy category, 102 about diseases category, and 42 about biological sciences category. Based on these semantic information,we built the neoplasm domain ontology. By now,this ontology includs : Classes 131,Object Properties 13.ConclusionsIn this paper we has attempted to built the neoplasm domain ontology based on text mining. And it has been verified that the concept modle of ontology can effectively express and reasonable organize the semantic information. With the further reseach of semantic web , Ontology in the field of text mining will play a greater role.