Prediction of Lung Cancer at Early Stage Using Correlation Analysis and Regression Modelling

In this paper we have developed a machine learning algorithm to detect the lung cancer depending on the symptoms. Using the various regression algorithm of machine learning, we have detected the lung cancer. We have compared the different regression algorithm and found the accuracy among them in predicting the lung cancer by considering the various symptoms like age, gender, and chest pain, shortness of breath, alcohol consumption, chronic disease, swallowing difficulty, anxiety and peer pressure. The regression algorithm like linear algorithm, polynomial regression, logistic regression, logarithmic regression and multiple regression are used to predict the lung cancer and found out the accuracy in predicting the lung cancer. The accuracy in predicting lung cancer using multiple regression is 96% which is more when compared to the other regression. The correlation between the various symptoms and lung cancer is also found out by finding the r square value using different regression machine learning algorithm. From the r square value that is obtained from various algorithm it's identified the lung cancer depends on major symptom like chronic disease.