• K. Lalithamani
• Dr.A. Punitha

Oral cancer analysis and its prediction tends to be a crucial area of research when concerning the safety of human life. Various Machine Learning (ML) techniques are adopted for analyzing and predicting the cancer data. The accuracy attained by predicted result is stated within average % by various researches. But by utilizing ML techniques, accuracy of maximum % can be obtained on UCI (Unique Client Identifier) medical oral cancer based dataset. The most significant factor in the diagnosis of any disease is the ‘Accuracy’. Therefore, it becomes mandatory to choose best suitable elements for producing optimal accuracy. The main objective of this paper is, to accurately predict the occurrence of oral cancer using minimum number of attributes. Altogether 25 attributes are detected for the prediction of oral cancer staging. The integrated diagnostic model with Principal Component Analysis (PCA) feature selection method is adopted here. With the help PCA feature selection algorithm 25 attributes are being cut down to only 10. Enhanced-Support Vector Machine (E-SVM) algorithm is proposed in this paper for oral cancer detection and identification in medical sector by employing ML approaches. E-SVMs are basically supervised learning models with associated learning algorithms that examines data which is being utilized for the purpose of classification and regression analysis. The components included in the proposed model are Data collection, Pre-processing, Feature selection via PCA, Clustering, Classification and Identification. The techniques of Data Mining (DM) aids in extracting significant patterns and relations from such data. Moreover, the ML and DM are effective techniques to detect and predict oral cancer at an early stage.

E-SVM, Oral Cancer, Data Mining, Machine Learning, Feature Selection via PCA, Clustering and Classification.

"ORAL CANCER DIAGNOSIS BASED ON FEATURE SELECTION AND CLASSIFICATION USING E-SVM ALGORITHM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.6, Issue 4, page no.247-252, April-2019, Available :http://www.jetir.org/papers/JETIR1904K35.pdf

"ORAL CANCER DIAGNOSIS BASED ON FEATURE SELECTION AND CLASSIFICATION USING E-SVM ALGORITHM", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.6, Issue 4, page no. pp247-252, April-2019, Available at : http://www.jetir.org/papers/JETIR1904K35.pdf