A New Feature Selection Method for Oral Cancer Using Data Mining Techniques

Abstract

The word cancer is used basically for more than 1000 different diseases including malignant tumours of different sites. Common to all forms of the disease is the failure of the mechanisms that regulate normal cell growth, explosion and cell death. Ultimately, there is evolution of the resulting tumour from mild to severe abnormality, with incursion of adjoining tissues and, ultimately, spread to other areas of the body. The primary risk factor for evolving oral cancer is tobacco use. Smoking cigarettes, cigars, and pipes all increase risk of oral cancer. Smokeless tobacco, also called "dip" or "chew," also enhance the risk. Alcohol consumption is another habit that is strongly associated with the growth of oral cancer. This paper uses data mining technology such as classification and prediction to identify oral cancer. Apriori algorithm is the innovation algorithm of Boolean association rules of mining frequent item sets. The datamining methods and techniques will be discovered to identify the suitable methods and techniques for efficient classification of data. The data mining techniques are effectively used to extract meaningful relationships from the data. Genetic algorithm were applied to association and classification techniques.