Documents classification by using ontology reasoning and similarity measure
Jun Fang, Lei Guo, Yue Niu · 2010 Seventh International Conference on Fuzzy Systems and Knowledge Discovery · 2010
Ontology-based documents classification method is introduced to solve the problem of classifier training and not considering semantic relations between words in traditional Machine Learning algorithms. However, previous work on ontology-based documents classification have some drawbacks on precision and run-time performance. In order to solve these problems, this paper proposes a novel ontology-based documents classification method by using ontology reasoning and similarity measure. Firstly, weighted terms set are extracted from documents, and categories are represented by ontologies; then the lowest concepts for each ontology is computed by using ontology reasoning techniques; next similarity score between documents and ontology is computed by using Google Distance measure; finally, web documents are assigned to categories according to the similarity score. Experimental results show our method is effective when comparing with the current ontology-based classification method, especially in the delicate classification evaluation, and the run-time performance is also better.