Call For Paper

Volume 10 Issue 3

March 2024

Submit Paper Here
Download Paper Format
Copyright Form
NEWS & UPDATES
News for Authors:

We have started accepting articles by online means directly through website. Its our humble request to all the researchers to go and check the new method of article submission on below link: Submit Manuscript

Follow us on Social Media:

Dear Researchers, to get in touch with the recent developments in the technology and research and to gain free knowledge like , share and follow us on various social media. Facebook

title

CREDIT CARD FRAUD DETECTION USING FREQUENT PATTERN MINING USING FP-MODIFIED TREE AND APRIORI GROWTH

Author(s):

Vipinkumar Choudhary

Keywords:

Abstract

The last few years have seen a major increase in the number of credit card users. This has brought into limelight the need to study users’ access patterns to bring to notice the usage patterns of fraudulent users. Data mining is one such field which is aimed at improving the user’s access experience and make it more secure. The last few years have seen many different techniques commonly referred to as “Association mining” by in the field of research. The previous works in this field have mostly been concentrated at Apriori and FP growth algorithms or their variants. We considered here two of the variants of Apriori and FP Growth algorithm which are Apriori growth and FP Split tree respectively. We take these as our base algorithms and create a hybrid of these two algorithms which combines the positive effects of each to bring forth an algorithm that has better performance than both the parents. The Apriori growth algorithm has a very highly time complex method for candidate generation as it involves multiple iteration of the database. But the method of generating frequent sets in Apriori growth is effective. On the other hand, the FP Split algorithm further improves the candidate generation method of FP Split tree by avoiding the recursive Subtree generation involved in FP Tree algorithm, but the method to generate frequent sets is very much the same as FP Tree algorithm. The previous works in this field have mostly been concentrated at Apriori and FP growth algorithms or their variants. The experimental results obtained by applying the algorithm on access logs from UCSD DataMining Contest 2009 Dataset (anonymous and imbalanced), will confirm the effectiveness of our algorithm in comparison to the two contemporaries. The results achieved by implementing the proposed technique in Java language confirmed the effectiveness of this methodology. The results were way better than its predecessor FP Tree algorithm and Apriori technique. The algorithms were implemented for various support levels.

Other Details

Paper ID: IJSARTV
Published in: Volume : 5, Issue : 9
Publication Date: 9/8/2019

Article Preview




Download Article