This paper is published in Volume-4, Issue-2, 2018
Area
Datamining
Author
K. Yamunadevi
Co-authors
R. Nagaraj
Org/Univ
Kaamadhenu Arts and Science College, Sathayamangalam, Tamil Nadu, India
Pub. Date
06 March, 2018
Paper ID
V4I2-1137
Publisher
Keywords
Datamining, Cancer, Information Gain(IG), Optimization Algorithm

Citationsacebook

IEEE
K. Yamunadevi, R. Nagaraj. An Optimized Classification of Human Cancer Disease for Gene Expression Data, International Journal of Advance Research, Ideas and Innovations in Technology, www.IJARIIT.com.

APA
K. Yamunadevi, R. Nagaraj (2018). An Optimized Classification of Human Cancer Disease for Gene Expression Data. International Journal of Advance Research, Ideas and Innovations in Technology, 4(2) www.IJARIIT.com.

MLA
K. Yamunadevi, R. Nagaraj. "An Optimized Classification of Human Cancer Disease for Gene Expression Data." International Journal of Advance Research, Ideas and Innovations in Technology 4.2 (2018). www.IJARIIT.com.

Abstract

Classification of cancer is determined the appropriate treatment and helps to determine the prognosis. The Existing System was presented to classify the human cancer diseases predicted on the gene expression profiles. The existing approach is initially, the Information Gain (IG) is utilized for feature selection and Genetic Algorithm (GA) is used for feature reduction. Finally the Genetic program is utilized for classifying the types of human cancer. However the existing approach has high computation time and required large amount of computational resource. To overcome this issue in this paper presented the Cuckoo Search (CS) optimization algorithm to optimize the threshold value of the features determined by the Information Gain (IG) and then Genetic programming (GP) is used for enhancing the performance of classifying the human cancer. The proposed cuckoo search approach is nature inspired behavior and breeding process of cuckoo bird’s optimization algorithm for generation of the global code book with one tuning parameter and it’s applicable for both linear and non linear problems. The performance of the proposed approach is evaluated in terms of Classification Accuracy, Specificity and Sensitivity.
Paper PDF