Efficient Feature Selection for Congenital Lungs Disorder in Human Fetus
K. Vimala1, D. Usha2
1K.Vimala, Research Scholar, Computer Science Department, Mother Teresa Women’s University Tamil Nadu, India.
2Dr. D. Usha, Assistant Professor, Computer Science Department, Mother Teresa Women’s University Tamil Nadu, India.
Manuscript received on July 20, 2019. | Revised Manuscript received on August 10, 2019. | Manuscript published on August 30, 2019. | PP: 2894-2897 | Volume-8 Issue-6, August 2019. | Retrieval Number: F8793088619/2019©BEIESP | DOI: 10.35940/ijeat.F8793.088619
Open Access | Ethics and Policies | Cite | Mendeley
© 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: Genetic disorders are one of major challenge in medical field, which is to be overcome in early stage, such that the patients can be diagnosed as soon as possible. This paper deals with fetal lungs disorder, as the initial process of the research proceeds with high dimensional dataset, where the fetal lungs dataset are preprocessed to check missing data or null criteria. Feature selection plays a major role here, where the denoised data are given to Principal component analysis, since the dataset was large in size; it was required to reduce the volume of data. Principal component analysis helps to reduce the redundancy in the data. The feature selection also provides minimum number of features, which is a pathway for performing the classification. Principal component analysis overcomes unrelated feature problem, increases the prediction accuracy level and decreases the computational overheads in classification. Efficiency of the feature selection is estimated using standard classification metrics.
Keywords: Genetic disorder, Lungs development, Preprocessing, Principal Component Analysis.