Applying Fuzzy Statistics and Clustering Analysis to Construct A Document Multi-Classification Model

Document classification is the procedure that a document is assigned into one of the predefined categories according to its content. In many cases, the content of a document may be involved in more than one issue; therefore it is reasonable to assign a document to more than one category. This kind of categorizing procedure is called multi-classification. Most of the document classification models have been designed for dealing with single-classification cases, therefore, a document multi-classification model is proposed. In this paper, the concept of fuzzy statistics analysis is used with clustering analysis to carry out the document multi-classification task. The results of experiment show that the performance of our multi-classification model is better.