An Innovative Model-Based Approach for Credit Card Fraud Detection Using Predictive Analysis and Logical Regression
S. Praveen Kumar1, A. Sahithi Choudary2
1S. Praveen Kumar, Assistant Professor, Department of Information Technology, GITAM Institute of Technology, GITAM (Deemed to be)University, Visakhapatnam (Andhra Pradesh), India.
2A. Sahithi Choudary, Department of Information Technology, GITAM Institute of Technology, GITAM (Deemed to be)University, Visakhapatnam (Andhra Pradesh), India.
Manuscript received on 18 April 2019 | Revised Manuscript received on 25 April 2019 | Manuscript published on 30 April 2019 | PP: 1926-1931 | Volume-8 Issue-4, April 2019 | Retrieval Number: D7020048419/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© 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: The development of information technology and advancements in communication channels has resulted in increasing fraud throughout the world and immense monetary losses. The objective of fraud detection frameworks is to check each exchange for the likelihood of being false and to recognize fraudulent ones as fast as possible after the fraudster has started to execute a fraudulent transaction, paying little mind to the prevention mechanisms. For this purpose, we utilize a steady foolproof 5 stage verification model with, predictive analysis, logistic regression, outlier model, custom rule management and global profiling. A predictive (LVQ) algorithm alongside logistic regression would improve credit card fraud detection. The benchmark Kaggle dataset is used. The outcomes portray a convincing decrease in credit card frauds.
Keywords: Credit Card Fraud Detection, Logistic Regression, Classification, Kaggle.
Scope of the Article: Classification