DOIONLINE

DOIONLINE NO - IJACEN-IRAJ-DOIONLNE-7621

Publish In
International Journal of Advance Computational Engineering and Networking (IJACEN)-IJACEN
Journal Home
Volume Issue
Issue
Volume-5, Issue-4  ( Apr, 2017 )
Paper Title
Artificial Activation System the Enzymatic Model for Classification of Imbalanced Data
Author Name
Anita Kushwaha
Affilition
Computer Science and Engineering Department Birla Institute of Technology, MESRA, Ranchi
Pages
5-17
Abstract
Imbalanced Dataset is a very common problem in classification of data. In supervised learning many techniques have been developed to tackle the problem of imbalanced training sets. Such techniques have been divided into two groups: at algorithm level and at the data level. Data level groups emphasized are those that try to balance the training sets by reducing the larger class through elimination of samples or increasing the smaller ones by constructing new samples known as Under sampling and Over sampling respectively. This paper proposes a new hybrid method for the classification of imbalanced datasets through construction of new samples using the Synthetic Minority Over sampling technique together with the application of a new technique Enzyme-computation called Artificial Activation System. The proposed method Enzyme-computation has been comparatively studied, validated and supported by an experimental study and shows good results. Keywords- Imbalanced Datasets, Oversampling, Under Sampling, rough set theory, Enzyme-computation model
  View Paper