Analysis of Oral Cancer Prediction with Pairwise Preprocessing Techniques using Hybrid Feature Selection and Ensemble Classification
Mumtazimah Mohamad1, Nurul Athirah Rozlan2, Fatihah Mohd3
1Mumtazimah Mohamad, Faculty Informatics and Computing, Universiti Sultan Zainal Abidin, Besut, Malaysia.
2Nurul Athirah Rozlan, Faculty Informatics and Computing, Universiti Sultan Zainal Abidin, Besut, Malaysia.
3Fatihah Mohd, School of Informatics and Applied Mathematics, Universiti Malaysia Terengganu, K. Terengganu, Malaysia.
Manuscript received on 16 February 2019 | Revised Manuscript received on 07 March 2019 | Manuscript Published on 08 June 2019 | PP: 605-611 | Volume-7 Issue-5S4, February 2019 | Retrieval Number: E11270275S419/19©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: Class imbalance is one of main problem in data mining field that can prompt to misclassification. Data are said to be imbalanced if the classes instances are not appearing similarly. Despite the fact that the sample of the dominant class and their appropriate classification are vital to classifier, oral cancer is analyzed by depending on the minority class tests. Numerous classification learning algorithms have low prescient precision for the rare class. Additionally, majority of the classification algorithms concern on the classification of significant major sample while overlooking the minority class. Misclassification resulted to non-cancerous and the cancerous patients pay expansion time and cost. In this research study, an examination of imbalanced classification issue on oral cancer prediction will be thoroughly performed. This investigation utilizes crossover approach of SMOTE and Random Undersampling and mix of feature selection strategies. The proposed algorithm is expected to gives better class imbalance solution and better performance in classification of oral cancer prediction.
Keywords: Class Imbalance, Data Preprocessing Techniques, Ensemble Algorithm, Feature Selection.
Scope of the Article: Regression and Prediction