Document Classification Using the Context of Terms

ABSTRACT One of the limitations of BOW method is that each term is recognized only by its form, failing to represent the term’s meaning or thematic background. To overcome the limitation, different profiles for each term were defined by thematic categ ories depending on contextual characteristics. In this study, a specific term was used as a c lassification feature based on its meaning or thematic background through the process of compa ring the context in those profiles with the occurrences in an actual document. The experi ment was conducted in three phases; term weighting, ensemble classifier implementation, and feature selection. The classification performance was enhanced in all the phases with the ensemble classifier showing the highest performance score. Also, the outcome showed that the proposed method was effective in reducing the performance bias caused by the total number of learning documents.키워드: 자동분류, 문맥프로파일, 용어가중치, 분류기 결합, 자질선정document classification, context profile, term weighting, ensemble classifier, feature selection