Detection of Oral Cancer Using Deep Neural Based Adaptive Fuzzy System in Data Mining Techniques
K. Lalithamani1, A. Punitha2
1K. Lalithamani, Research Scholar, Bharathiar University, Coimbatore (Tamil Nadu), India.
2Dr. A. Punitha, Research Guide, Bharathiar University, Coimbatore (Tamil Nadu), India.
Manuscript received on 24 April 2019 | Revised Manuscript received on 02 May 2019 | Manuscript Published on 08 May 2019 | PP: 397-404 | Volume-7 Issue-5S3 February 2019 | Retrieval Number: E11720275S19/19©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Cancer has the highest growth rate among all diseases globally. Oral cancer is one of the most dangerous cancer which affects and originates from the oral cavity and neck. Overuse of tobacco and smoking cigarettes are the primary risk factor for developing oral cancer. Another habit which has a strong association with oral cancer is the consumption of alcohol. A large number of patient deaths were recorded from oral cancer as a result of lack of its identification and late treatment. Researchers in the medical community are making an effort to provide a system for effective diagnosis and prevention of the serious disease. In the present research, oral cancer patients can be identified through the use of data mining technology which includes detection, classification and clustering. A Deep Neural Based Adaptive Fuzzy System (DNAFS) is proposed in this paper which uses machine learning for the detection and identification of oral cancer. The two techniques which are part of DNAFS are based on fuzzy logic and neural networks. DNAFS and methods for data mining are explored for the identification of suitable techniques which are helpful in classifying data efficiently. The stages included in the proposed mechanism include data collection, pre processing, Fuzzy C-Means for clustering data, feature selection, classification and identification. Meaningful relationships can be extracted effectively from data using data mining techniques. About 96.29 % accurate results are available from experiments. There is less than 5 ms incidence of error in the result. The datasets are required to be investigated further in daily clinical practice.
Keywords: Data Mining, Feature Selection, Machine Learning, Fuzzy C-Means, Oral Cancer, Medical Data Set.
Scope of the Article: Fuzzy Logics