[3Rin4] Interactive 1

[3Rin4-54] An Idea of a Rough Set Theory Based Document Classification System with RT method

〇Masaki Kurematsu¹ (1.Iwate Prefectural University)

Keywords:Rough Set Theory, Document Classification, RT Method

In this paper, I propose a Rough Set Theory based document classification system with RT method. First, the proposed system makes a decision table by combining the label of documents and terms extracted by the document frequency. After getting reduction from this decision table, it makes decision rules from lower approximation. Next, it makes samples for RT method by matching these decision rules to the decision table. In this step, it uses Satisfaction Index, Coverage Index ,Lift and Support as these rule’s weight. Finely, it makes the unit space based on samples. To identify an unlabeled document, it converts this document to sample data by the same way for making samples and get the distance between this document and the unit space. It says the class which has the shortest distance. In order to evaluate this approach, I implemented a prototype system and tried to classify labeled patent publications in Japanese with experts. This system had extracted some rules evaluated as useful by an expert, however, the accuracy is not good. Therefore, I try to improve this method based on the experimental result.