Classification Method for Imbalanced Data using Ensemble Learning System
Sunil Chandolu1, S.Prasad Babu Vagolu2

1Sunil Chandolu*, Dept. Of CS, GIS, GITAM University, India.
2S.Prasad Babu Vagolu, Dept of CS, GIS, GITAM University, India.
Manuscript received on January 10, 2020. | Revised Manuscript received on January 22, 2020. | Manuscript published on February 10, 2020. | PP: 1845-1848 | Volume-9 Issue-4, February 2020. | Retrieval Number: B6289129219/2020©BEIESP | DOI: 10.35940/ijitee.B6289.029420
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 (

Abstract: In this research, arrangement including imbalanced datasets has gotten extensive consideration. Generally, order calculations will, in general, anticipate that the majority of the approaching information has a place with the greater part class, bringing about the poor arrangement execution in the smaller number or part occasions, which are ordinarily of considerably more intrigue. In this paper, we propose a grouping based subset troupe learning strategy for taking care of class imbalanced issue. In the proposed methodology, first, new adjusted preparing datasets are delivered utilizing bunching based Under-inspecting, at that point, a further grouping of new training sets is performed by applying four calculations: Decision Tree, Naive Bayes, KNN and SVM, as the base algorithms in joined packing. A test investigation is completed over a wide scope of exceptionally imbalanced datasets. The outcomes acquired show that our technique can improve the irregularity order execution of uncommon and ordinary classes steadily what’s more, successfully. 
Keywords: Imbalanced Information, Classification, Clustering, Ensemble learning.
Scope of the Article: Clustering