ICA Learning Approach for Predicting RNA-Seq Data Using KNN and Decision Tree Classifiers

Abstract

Parasites such as malaria accept disturbed variation of life unit’s growth through uncountable stratospheres of mosquito vector. There are transcriptomes with distinct species, Ribonucleic acid sequencing (RNA-seq) is a predominant protein sequence expression process leading to increase genetic query identification. RNA-seq calculate records of gene expressions and imposes machine learning diagnostic enhancements techniques. Biological data learning approaches for analyses have been suggested by numerous researchers. An Independent Component Analysis (ICA) feature extraction dimensionality reduction algorithm is proposed to draw hidden components from RNA-seq dataset with high dimension, utilizing KNN and Decision Tree classification procedures to evaluate the performances, the study attained a performance metrics of 81.7.7% and 73.3% classification accuracy respectively.