The increasing amount of text information available online has led to the development of numerous different approaches for text classification. The most popular straightforward and effective method is the cosine similarity classifier. It enhances the performance of text document classification. It is paired with the predicted value that traditional classifiers, such the Mamdani fuzzy rule-based system, provide. Thus, the classifier performs better when the estimated value for a category is combined with similarities between test documents and that category. This method offers an efficient and successful way to categorize text documents. Furthermore, techniques for ascertaining the appropriate correlation between a group of terms in a document and its classification are also acquired.