Computational Identification of Lungs Cancer Causing Genes by Machine Learning (Ml) Classifiers

Abstract

Molarity rate is increasing day by day at all over the world among both genders due to the increasing rate of lung cancer. It is a dangerous disease and usually it starts when unrestrained growth of abnormal cells start growing in lungs. The early detection of this disease has been a major challenge in the past hence, to overcome this issue many detection techniques have introduced over the time. In last decade, many Machine Learning classifiers have developed and adopted for the detection of lungs cancer. In this study, we have utilized six ML classifier such as ‘Support Vector Machine ‘(SVM) ‘K-Nearest Neighbor’ (KNN), Adaboost, ‘Conventional Neural Network’ (CNN), Xgboost and Naive Bayes for the detection of lungs cancer causing genes. We have collected dataset from publicly available intoGene browser. This dataset consists of 2193 genes in which both tumor and non-tumor genes are included. To find, which classifier provide high accuracy of lungs cancer detection as well as lungs cancer causing genes, this study have used the above-mentioned ML classifiers and found that CNN proved to be the best classifier with 86 percent accuracy among all classifiers.