Provisional Research on Ensemble Learning Techniques for Card Fraud Detection
Pooja Pant1, Prakash Srivastava2, Ashutosh Gupta3
1Pooja Pant, Amity University, Noida (U.P), India.
2Dr. Prakash Srivastava, Amity University, Noida (U.P), India.
3Dr. Ashutosh Gupta, UPRTOU, Allahabad (U.P), India.
Manuscript received on 15 August 2019 | Revised Manuscript received on 27 August 2019 | Manuscript Published on 06 September 2019 | PP: 13-17 | Volume-8 Issue- 6S, August 2019 | Retrieval Number: F10030886S19/19©BEIESP | DOI: 10.35940/ijeat.F1003.0886S19
Open Access | Editorial and Publishing 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: Machine learning have revolutionized fraud detection in various domains like telecommunication and e-commerce. Global statistics shows how billions of dollars are lost because of card frauds every year and millions of people falling the victims. Fraud detection systems used for credit card fraud detection 2 decades ago are still being used because of the trust and stability they have provided for so long. With a number of academic research being done in fraud detection their effect on the financial industry has been minimum. Even with high prediction accuracy using machine learning approaches like deep learning and stack ensemble most of these research gets directly rejected by the industry. Our research objective is to highlight the reason of rejectection which are mostly ignored by the researchers and there adverse effect on the results.
Keywords: Ensemble Learning, Machine Learning, Fraud Detection.
Scope of the Article: Machine Learning