An Ensemble Model With Feature Selection Technique for Classification of Lung Cancer Disease

Abstract

Cancer is very serious and dangerous disease facing by many people in the world. Lung cancer is one of the most dangerous cancer types which directly affected to the human life. This disease can spread worldwide by uncontrolled cell growth in the tissues of the lung. An identification and classification of lung cancer is very necessary to diagnosis of lung cancer disease. In this paper we have analyzed the lung cancer prediction using data mining based classification algorithm such as J48, LMT, REP Tree, CART, Bayes Net, Naïve Bayes, SVM and its ensemble model. The proposed ensemble of Naïve Bayes and LMT gives better classification accuracy compare to others. We have also applied the various ranking based feature selection techniques on proposed ensemble model and achieved better classification accuracy with less number of features and less computational time.

An Approach to Diagnosis of Thyroid using Data Mining Techniques

“…The occurrence of this ailment might also sometimes result into hyperthyroidism or hypothyroidism. A hyperthyroidism is scenario; wherein large number of such hormones are produced which might in turn lead to an increase in weight of the human body [1]. On the other hand, presence of hypothyroidism is triggered by less production of such hormones which can eventually lead to other medical ailments and disorders such as inflammation and swelling.…”