Machine Learning Approaches for Lung Cancer Prediction

Abstract

Cancer is a long-term, exhausting disease that requires changes in all living conditions of the patient and his/her environment. Although there are regional variations in deaths from all causes in the world, it is in the 3rd rank. Lung cancer is among the most frequent cancer kinds worldwide, regardless of male or female. Cancer is a preventable disease. To prevent a disease, it is necessary to know its causes and avoid them. The use of tobacco and tobacco products is the main risk factor for all cancers, especially lung cancer. Early diagnosis of cancer is lifesaving. According to the Turkish Respiratory Research Association, 200,000 people are diagnosed with cancer every year in our country. With the accelerated developments in technologies and the digitalization of health services, a large amount of cancer data has been collected and this data has been used by many researchers, especially in low and middle-income countries, to reduce the cost of tests used to predict different cancer types and to predict different cancer types. This article is exploited various machine learning algorithms for predicting lung cancer. Experimental results show that random forest performed better by attaining 96.08% accuracy.