Improved Weighting in the Automated Texts Classification using Fuzzy Method

In the current digital landscape, particularly on the web, there is an increasing demand for improved methods of controlling and organizing written documents. Automatic text classification, a crucial aspect of text mining, plays a significant role in managing systems effectively. The accurate weighting of keywords greatly influences the outcomes of text-mining techniques. In this article, we present an enhancement to the TFIDF weighting method, incorporating two important factors: the distribution of keywords throughout the entire text and giving priority to the opening and conclusion paragraphs based on the fuzzy method. To assess the effectiveness of our proposed approach, we classified 500 texts into 5 distinct categories. Using the proposed method has led to the improvement of 4%, 3%, 13%, 2%, and 5% in each of the parameters F-measure, Recall, Precision, FP, and TP. The results demonstrate the advantages of our method, highlighting its superior performance.