A Fuzzy Similarity-Based Approach for Multi-label Document Classification

Multi-label document classification concerns the determination of categories in the situation where one document may belong to more than one category. In this paper we propose a fuzzy similarity-based approach for multi-label document classification. For a test document, the scores of its relevance to the classes are calculated based on a modified fuzzy similarity measure. The test document is then decided to belong to every class whose score passes a threshold. To make the system adaptive, we provide a heuristic approach to find a score threshold automatically for each class. Experimental results show that our proposed method is more effective and efficient than other existing methods.