Outlier Detection Using Oversampling PCA for Credit Card Fraud Detection
Journal Title: International Journal for Research in Applied Science and Engineering Technology (IJRASET) - Year 2017, Vol 5, Issue 5
Abstract
Credit card fraud detection is an important application of outlier detection in recent years. Many outlier detection techniques are available but they are working in batch mode, due to this those techniques are not applicable for applications where large amount of data is present. In this paper, a new credit card fraud detection method based on Oversampling Principal Component Analysis (osPCA) with low computation and memory requirements is presented. Principal Component Analysis (PCA) is a tool in data analysis for dimension reduction, it transforms high dimensional data into lower dimensions which contains maximum amount of information. Here we are using concept of oversampling on data and then input is given to PCA, therefore it is possible to detect credit card frauds from large amount of data. This technique is suitable for online applications which have memory or computation limitation.
Authors and Affiliations
Amruta D. Pawar, Seema A. Dongare, Amol L. Deokate, Harshal S. Sangle, Panchsheela V. Mokal
Keywords
Related Articles
- EP ID EP24156
- DOI -
- Views 304
- Downloads 12
Online Housing Society Management System
This paper presents the various methods in which we can manage the housing society by a system which is created using cloud. And makes the current situation in the society simple and efficient. As we know there is an in...
Review of Integrity Constraint in MySQL
For the semantic correctness of database records, Integrity constraints or data constraints are proper representations of invariant conditions. Constraints can be articulated in declarative languages such as data log, p...
Recent Trends in Solar Distillation
Water is the main source for sustaining the man on earth. Groundwater and reservoirs are the largest available sources of freshwater to fullfill needs of mankind. However, these sources do not always proved to be useful...
Prediction of Heart Disease Based on Decision Trees
The prediction of heart disease is most complicated task in the field of medical sciences which cannot be observed with a naked eye and comes instantly anywhere, anytime. So there arises a need to develop a decision sup...
Savoir-faire malware tracking down in Delay Tolerant Networks
With the universal presence of short-range connectivity technologies (e.g., Bluetooth and, more recently, Wi-Fi Direct) in the consumer electronics market, the delay tolerant-network (DTN) model is becoming a viable alt...