Hybridizing Gray Wolf Optimization (GWO) with Grasshopper Optimization Algorithm (GOA) for text feature selection and clustering

Abstract Text analysis in the field of text mining requires complex techniques for handling several text documents. Text clustering is among the most effective tactics in the field of text mining, machine recruitment and pattern recognition. Computers can start organizing a corpus document in certain organizational structure of conceptual clusters using reasonable text-clustering method. Informative and un-informational functionalities of the text documents contain noisy, inconsequential and superfluous features. The main method of finding a new subset of informative feats for each document is the unsupervised selection of text features. The functional selection technique has two aims: (1) maximize text clustering algorithm reliability, (2) minimize the number of uninformative traits. The proposed technique is that it produces a mature convergence rate and requires minimal computational time and is trapped in local minima in a low dimensional space. The text data is fed as the input and pre-processing steps are performed in the document. Next, the text feature selection is processed by selecting the local optima from the text document and then selecting the best global optima from local optimum using hybrid GWO–GOA.​ Furthermore, the selected optima are clustered using the Fuzzy c-means (FCM) clustering algorithm. This algorithm improves the reliability and minimizes the computational time cost. Eight datasets are used in the proposed algorithm and the performance is envisaged efficaciously. The evaluation metrics used for performing text feature selection and text clustering are accuracy, precision, recall, F-measure, sensitivity, specificity and show better quality when comparing with various other algorithms. When comparing with GWO, GOA and the proposed hybrid GWO–GOA algorithm, the proposed methodology reveals 87.6% of efficiency.