Lung Cancer Prediction Using Machine Learning Methodologies

Abstract

Lung cancer usually occurs in both men and women as a result of unmanageable lung cell growth. This constitutes a severe respiratory problem in both the inhalation and the exhalation of the chest. Cigarette smoking and tobacco smoke are the main contributors to lung cancer in the global health organisation. The death rate due to lung cancer in young and old people is rising day by day in comparison with other cancers. Although high-tech medical facilities for been well and efficient medical treatment are available, the mortality rate is not yet properly monitored. Therefore way earlier safety measures are extremely necessary in the initial stage so that symptoms and effects can be identified early in the process for proper diagnosis. Nowadays, machine learning has a big impact on the health sector as a result of its high computational capacity for early disease diagnosis with reliable data analysis. In our paper, we analysed various techniques for machine learning to classify available lung cancer information into a malignancy UCI process was developed. The input data are pre-empted and transformed into binary form followed by a well-known Weka classifiers technique to classify data from cancer to non-cancerous. The method of comparison demonstrate that the suggested RBF classifier had a high accuracy of 81.25 percent and was regarded as the efficient classifiers method for prognostication of Lung cancer.