Optimised fuzzy C-means clustering based cluster indexing towards incremental learning integrated text categorisation

Abstract

Text classification is the most significant task in the data retrieval process through classifying text into various groups depending on the document's content. The quick progression of electronic documents may produce various issues, such as unstructured data, which requires more effort and time for searching appropriate documents. Text categorisation has high importance in information retrieval and processing, wherein unstructured documents are arranged into a predefined group. In addition, incredible growth in online documents obtains ability with the development of the internet needs a highly precise and effective retrieval approach. Thus, in this paper, Dingo Monarch butterfly optimisation (DMBO) approach and Tversky index-based indexing are developed for incremental learning-enabled text categorisation. Moreover, text categorisation is done based on incremental learning along with a Bayesian classifier. This text classification approach achieved better performance with a precision of 0.9136, recall of 0.9173, F-measure of 0.9051, and accuracy of 0.8461.