Early Lung Cancer Prediction using Correlation and Regression
K Sivanagireddy, Yerram Srinivas, S. Sri Nandhini Kowsalya, S. Sivasankari, J. Surendiran, R.G. Vidhya
Abstract
In this study, we created a machine learning approach for symptom-based diagnosis of lung cancer. Lung cancer detection was accomplished using a number of machine learning regression strategies. By assessing the efficacy of many regression algorithms in predicting lung cancer, we can better understand the risk factors and symptoms associated with this illness. Lung cancer predictions and evaluations are made using regression methods such the linear algorithm, polynomial regression, logistic regression, logarithmic regression, and multiple regression. Compared to other regression approaches, multiple regression has a 96% higher accuracy in predicting lung cancer. The r-squared value may be calculated using a number of regression machine learning algorithms, making it possible to evaluate the association between the symptoms and lung cancer. Several algorithms calculate a <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{r}$</tex> squared value based on key symptoms, such as long-term illness, to diagnose lung cancer.