Incorporating Artificial Fish Swarm in Ensemble Classification Framework for Recurrence Prediction of Cervical Cancer

Abstract

IoT has facilitated predominant advancements in cancer research by incorporating Artificial intelligence (AI) that enables human decision-makers to achieve a better decision. Recently, Least Absolute Shrinkage and Selection Operator (LASSO) classifier has launched in predicting recurrence cancer genes in the cervix. At the initial phase, the recurrence gene expression of lncRNA is collected from Geo Datasets. Secondly, data imputation, accomplished with Mode and Mean Missing method (MMM-DI). Thirdly, feature selection is compassed using Hilbert-Schmidt independence criterion with Diversity based Artificial Fish Swarm (HSDAFS). In the HSDA.FS algorithm, the diversity parameter is added based on the gene value, and their risk score of the lncRNAs is computed using the Artificial intelligence (AI) technique. Finally, recurrence prediction, an ENSemble Classification Framework (ENSCF), is proposed based on recurrent neural networks. The prognostic factor is computed with a risk score of nine lncRNA signatures for 300 samples taken from GSE44001. The Chi-Square method has been used to obtain statistical results. The survival of the patient with recurrence cervical cancer is shown using the proposed model.