currently, most text categorization based on semantics theory only stay in the study, with very few specific areas to study. Therefore, the need for further research and development needed for certain areas of text classification.We use the classification of related technologies to solve the problem of the lack of semantics, we will combine machine learning algorithm and the concept vector model,this paper will introduce a concept representation of text which makes use of domain ontology knowledge to obtain relationships between words in the text, and eventually form a concept vector space model as well as a simple vector distance classification will apply it to realize domain text categorization.
展开▼