Credit Card Fraud Detection using Machine Learning
Gautam Kumar1, Shivanesh Kumar2, A Arul Prakash3
1Gautam Kumar*, Student, Department of School of Computer Science and Engineering, Galgotias University, Greater Noida, India
2Shivanesh Kumar, Student, Department of School of Computer Science and Engineering, Galgotias University, Greater Noida, India
3A Arul Prakash, Assistant Professor, Department of School of Computer Science and Engineering, Galgotias University, Greater Noida, India
Manuscript received on March 10, 2021. | Revised Manuscript received on March 16, 2021. | Manuscript published on April 30, 2021. | PP: 124-126 | Volume-10 Issue-4, April 2021. | Retrieval Number: 100.1/ijeat.D23440410421 | DOI: 10.35940/ijeat.D2344.0410421
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: Now a days credit card plays a very important role in the lives of the human being. It becomes an important part of the businessman, global activities and many more. Even using credit cards give us a most widely used of benefits when it used with the responsibility and carefully, and very small credit and financial harm is also caused by fraudulent activities or transactions. There are a lot of techniques are given to encounter the scope in credit. In spite of, whatever the methods are used they have the same goal of clog the card fraud and each one has its own advantage, drawback and the characteristics too. The deficiency and the good of the credit card detection-methodologies are description and dissimilarity. Moreover, a taxonomy of reference techniques are classified in two fraud-detection perspective, as misuse (supervised) and absurdity (unsupervised) is given. Again, a taxonomy of methods is presented supported caliber to process the categorical and numerical datasets. Other kind of datasets are made in the literature then mentioned and sorted in real and club into the group of the data and therefore the dominant and customary attributes are removed for prosecute application. Consequently, for the new researches, the issues for credit card fraud-detection are described as per the recommendations.
Keywords: Purchasing, Clustering, Datasets, Random Forest, Naïve Bayes Classifie.r.
Scope of the Article: Machine Learning