Feature Selection for Lung Cancer Detection Using SVM Based Recursive Feature Elimination Method

Author(s):  
Kesav Kancherla ◽  
Srinivas Mukkamala

Lung Cancer is the most general type of disease in theworld ofcancer. It affects the lungs of the human body. So, the prediction of lung cancer at its earlier stage is difficult. It is the deadliest cause of death in both men and women. Its symptoms are harder to recognize in the initial stages.Machine learning algorithms have made the prediction and detection of lung cancereasier. Chi-square is used for feature selection to select the relevant features in the lung cancer dataset. Different Machine Learning algorithms are used to predict Lung Cancer.The algorithmsutilized in the proposed work are SVM and Random Forest. We have compared these algorithms with and without feature selection (Chi-square). SVM is identified as the best algorithm in the proposed work due to its accuracy and less execution time for detecting the model. The key objective of this paper is to enhance the accuracy and reduce the execution time of the classifier.


Lung cancer is the foremost cause of cancer-related deaths world-wide [1]. It affects 100,000 Americans of the smoking population every year of all age groups, particularly those above 50 years of the smoking population [2]. In India, 51,000 lung cancer deaths were reported in 2012, which include 41,000 men and 10,000 women [3]. It is the leading cause of cancer deaths in men; however, in women, it ranked ninth among all cancerous deaths [4]. It is possible to detect the lung cancer at a very early stage, providing a much higher chance of survival for the patients.


Sign in / Sign up

Export Citation Format

Share Document