ISSN (Online) : 2456 - 0774

Email : ijasret@gmail.com

ISSN (Online) 2456 - 0774


UTILIZING PRODUCT FEATURES FOR FRAUD DETECTION ON E-COMMERCE PLATFORMS IN BIG DATA TRANSACTIONS

Abstract

Abstract:- - Fraudulent transactions are a seriousconcern for e-commerce websites and other platforms that are operating theirbusinesses on the web. With the advent and growth of Big Data technology,consumers who buy products online always assess the vendors or suppliers as perthe ratings and reputation suggested by the e-commerce platform. The cause ormotivation for the vendors to pursue high ratings and reputation scores forthem on the e-commerce platform is that high ratings and positive commentsabout the products sold by them would fetch them high profits. The fraudulentvendors try to attain high reputation scores and thereby attract more customersto buy their products only. It is very much important for e-commerce websitesto identify such imposters and identify fake reputation information as notidentifying the fraudulent vendors would lead to the loss of business andreputation of the e-commerce platform itself. The e-commerce platforms nowadaysare attempting to curb ongoing and growing issue by employing data miningmechanisms. With the advent of the Internet of Things (IoT) and Big Data has animportant role to play in the economic growth in various domains. It supportsthe organizations and improves their decision-making capabilities by analyzingtheir operational data. It also helps the e-commerce platforms by providingonline customers with an impartial and strong reputation system, therebyimproving their online shopping experience. The objective of this technicalpaper is to present a conceptual framework to bring out the characteristics offake transactions along with individual and transaction-related pointers.Product type and Product nature are two features to which are used to identifyfraudulent transactions. These two features help in improving the accuracy offake reputation detection. A dataset from the real-world is used to validatethe effectiveness and accuracy of the fraud detection model which helps inidentifying the fraudulent vendors from the genuine ones.

Keywords:-E-business,Fraud Detection, Reputation System, SNA, K core


Full Text PDF

IMPORTANT DATES 

Submit paper at ijasret@gmail.com

Paper Submission Open For March 2024
UGC indexed in (Old UGC) 2017
Last date for paper submission 30th March, 2024
Deadline Submit Paper any time
Publication of Paper Within 01-02 Days after completing all the formalities
Paper Submission Open For Publication /online Conference 
Publication Fees  
Free for PR Students