Nonlinear transformation of term frequencies for term weighting in text categorization
Engineering Applications of Artificial Intelligence
| Hi-index | 0.00 |
This paper proposes a new text categorization method based on the similar rough set and the fuzzy cognitive map (FCM). The method replaces casual relation with correlation relation in the fuzzy cognitive map, called correlation-FCM, for constructing text classifiers. Text categorization can be implemented by reasoning approaches based on fuzzy cognitive maps which include weights and correlation degrees. We use similar rough set to generate an efficient representation of documents and apply correlation-FCM for training text classifiers. We apply this proposed method to the WebKB benchmark datasets and the results prove that the method is a good choice for applications with a limited amount of labeled training data. We also demonstrate the effect of changing training size on the classification performance of the learners.