Detection of Lung Cancer in Smokers and Non-Smokers by Applying Data Mining Techniques

Abstract

Objective: The data of medical science is increasing rapidly from the last few years. Extensive information related to any disease and its symptom can be extracted, such information can be used for early detection of diseases and to overcome the disease in better way. In this research we are focusing on lung cancer detection at earlier stage. Methods/Statistical Analysis: To discover the meaningful knowledge from the dataset for the health professionals to find the disease at earlier stage with low cost. In this regard, personal characteristics like age, gender, socio economic factors were considered by applying C5.0 algorithm for the development of model. Findings: We found that in Pakistan, lung cancer ratio is increasing in Non-Smoker, previously occurrence ratio was 20% and 80% in Non-Smoker and Smokers respectively but in this research ratio is 22% and 78%. Dataset was taken from SKMCH & RC database. Application/Improvements: Results of this research will be useful for the oncologists; research will be extended in near future on different datasets of lung cancer to diagnose the reasons of lung cancer in non-smokers.