Please use this identifier to cite or link to this item:
http://hdl.handle.net/2080/3366
Title: | Web service based credit card fraud detection byapplying machine learning techniques |
Authors: | Prusti, Debachudamani Santanu Kumar, Rath |
Keywords: | Credit card Majority voting Fraud pattern, Classification tech-nique, Prediction Accuracy. Web service |
Issue Date: | 2019 |
Citation: | Tencon 2019 ,17 - 20 October 2019 |
Abstract: | Financial fraud in the credit card transaction environment refers to the unauthorized use of card transactions through multiple platforms by stealing the user credentials to gain advantages fraudulently. In order to curb this problem,an accurately financial fraud detection method is necessary to implement; since this type of fraud causes financial loss with security breach. It is observed in literature that application of machine learning techniques yields satisfactory result for fraud detection in the credit card operations and hence, it has been an important area of research in the domain of digital transactions.In this study, we have proposed the application of commonly applied classification techniques such as Decision tree (DT),Extreme learning machine (ELM), K-Nearest neighbor (K-NN), Multi layer Perceptron (MLP) and Support Vector Machine(SVM) to find the accuracy for the fraud detection. Further,we have proposed a model by hybridizing DT, SVM and KNN models, in which the prediction accuracy has improved significantly. Apart from this, two web services Simple Object Access Protocol (SOAP) and Representational State Transfer(REST) have been applied in this study for efficient exchange of data across heterogeneous platform. |
Description: | Copyright of this document is with the conference publisher |
URI: | http://hdl.handle.net/2080/3366 |
Appears in Collections: | Conference Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Paper presented at TENCON-2019.pdf | conference paper(Main) | 1.99 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.