Comprehensive Assessment of Imbalanced Data Classification
Smita Nirkhi1, Shashikant Patil2

1Smita Nirkhi1, Computer Engineering, Shri Ramdeobaba College of Engineering & Management, Nagpur.
2Shashikant Patil*, EXTC Department,SVKMs NMIMS Shirpur Campus.

Manuscript received on March 30, 2020. | Revised Manuscript received on April 05, 2020. | Manuscript published on April 30, 2020. | PP: 1426-1431 | Volume-9 Issue-4, April 2020. | Retrieval Number: D7349049420/2020©BEIESP | DOI: 10.35940/ijeat.D7349.049420
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: This is an attempt to address the various challenges opportunities and scope for formulating and designing new procedure in imbalanced classification problem which poses a challenge to a predictive modelling as many of AI ML n DL algorithms which are extensively used for classification are always designed from the perspective of with majority of focus on assuming equal number of examples for a class. It leads to poor efficiency and performance especially in minority class. As Minority class is always very crucial and sensitive to classification errors and also its utmost important in imbalanced classification. This chapter discusses addresses and gives novel as well as deep insights with unequal distribution of classes in training datasets. Largely real time and real world classifications are comprising imbalanced distribution so need specialized techniques for more challenging and sophisticated models with minimal errors and improved performance.
Keywords: Imbalanced Data; Class imbalance, Machine Learning; Algorithms; Data Mining .