Anomaly Detection Algorithms in Financial Data
Abhisu Jain1, Mayank Arora2, Anoushka Mehra3, Aviva Munshi4

1Abhisu Jain*, Department of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India.
2Mayank Arora, Department of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India.
3Anoushka Mehra, Department of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India.
4Aviva Munshi, Department of Computer Science and Engineering, Vellore Institute of Technology, Vellore (Tamil Nadu), India.

Manuscript received on May 21, 2021. | Revised Manuscript received on May 10, 2021. | Manuscript published on June 30, 2021. | PP: 76-89 | Volume-10 Issue-5, June 2021. | Retrieval Number:  100.1/ijeat.E25980610521 | DOI: 10.35940/ijeat.E2569.0610521
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 (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: The main aim of this project is to understand and apply the separate approach to classify fraudulent transactions in a database using the Isolation forest algorithm and LOF algorithm instead of the generic Random Forest approach. The model will be able to identify transactions with greater accuracy and we will work towards a more optimal solution by comparing both approaches. The problem of detecting credit card fraud involves modelling past credit card purchases with the perception of those that turned out to be fraud. Then, this model is used to determine whether or not a new transaction is fraudulent. The objective of the project here is to identify 100% of the fraudulent transactions while mitigating the incorrect classifications offraud.
Keywords: Isolation Forest, Local Outlier, Credit Card, Anomaly Detection