Improving Documents Classification with Semantic Features
Bai Rujiang, Junhua Liao · 2009
Successful text classification is highly dependent on the representations used. Currently, most approaches to text classification adopt the `bag-of-words' document representation approach, where the frequency of occurrence of each word is considered as the most important feature, but this method ignores important semantic relationships between key terms. In this paper, we proposed a system that uses ontologies and Natural Language Processing techniques to index texts. Traditional BOW matrix is replaced by "Bag of Concepts" (BOC). For this purpose, we developed fully automated methods for mapping keywords to their corresponding ontology concepts. Support Vector Machine a successful machine learning technique is used for classification. Experimental results shows that our proposed method dose improve text classification performance significantly.