Text feature selection and clustering Based on hybridized gray wolf Optimization grasshopper optimization Algorithm

Abstract

Data mining refers to the extraction of knowledge from large data. It newlineextracts patterns and develops relationships from various unrefined data and newlinemakes use of diverse tools of investigation. The different websites along with the newlineother types of computerized implementation have been offering plenty of text newlineinformation. With regard to efficiently making use of various details, text newlineinformation has to be divided owing to the growth in the use of such newlineapplications. There are transcripts that have been categorized according to their newlineprovenance with plenty of information that is in the form of saving texts on newlinedigital documents. The strategies of text clustering have been used extensively newlinefor categorizing the formation of texts. Text clusters are further used along with newlinetext mining implementation, text classification, and retrieval of text information. newlineFor all other areas of text mining, they may be specific to the field of clustering newlineof text documents which can be a hurdle to the techniques of text analysis. newlineThe main purpose of the feature selection process is to remove newlineunwanted information from a feature set. In the initial stage of work, the Fuzzy newlineC Means (FCM) and K-Means clustering, Correlation Based Feature Selection newline(CFS) and Mutual Information (MI) based feature selection method and Particle newlineSwarm Optimization (PSO) algorithm are evaluated. FCM clustering is newlinefrequently used in pattern recognition; it allows one piece of data to belong to newlinetwo or more clusters. The K-means algorithm, a widely used flat clustering newlinealgorithm, minimizes the average distance of documents from their cluster newlinecenters. newline