Global Journal of Commerce & Management Perspective
Open Access

Mechanism for Employing Logical Regression for Detecting Corruption at the Point of Sale on e-commerce Platforms

Daveed Smeeth^{* *}Correspondence: Daveed Smeeth, Department of Economics, Harvard University, Cambridge Street, Cambridge, USA, Email:

Author info »

Description

Electronic commerce has shown to be beneficial for numerous enterprises. As long as there is an internet connection and a device available for internet browsing, it has made it possible for businesses and individuals to conduct business digitally and enjoy variety. Across the nation, a number of administrations have progressively embraced electronic payment. In addition, the Nigerian government has actively encouraged the country's citizens to embrace e-commerce and establish a cashless economy. Fraud through this channel, both real and attempted, is growing along with e-commerce. The Nigerian Central Bank predicted that by 2021, electronic fraud would amount to trillions of Naira.

The aim of this study was to use logistic regression as a tool for decision-making to identify fraud at the point of sale in ecommerce platforms. This study's primary contribution is a methodology designed to identify fraud on e-commerce platforms at the point of sale. The result has a 97.8% accuracy rate. The findings of this study will give important decisionmakers at e-commerce company’s knowledge about fraud trends on their e-commerce platforms, allowing them to move quickly to thwart these fraudulent attempts. It is important to conduct additional studies in developing nations. Modern technology has brought about considerable changes to the financial system, making it more dynamic and ushering in an era of electronic payments. These advancements have also altered traditional payment methods. Any kind of transaction made over an electronic channel that doesn't include the usage of cash or checks is referred to as an electronic payment. The emergence of electronic commerce (ecommerce) has been made possible by electronic payment, which has also enhanced consumer preferences, usability, cost, security, acceptability, and relevancy. All of these factors contribute to the success of ecommerce payment systems.

The emergence of e-commerce has made it possible for many businesses to grow across national borders in order to access a larger market. Thanks to e-commerce, businesses can now operate year-round. Customers should expect convenience, speed, variety, price comparison, and much more from ecommerce. Customers are spared from having to go to actual establishments each time they want to make a transaction. As evidenced by the success of businesses from different companies and others, consumers have embraced online shopping over traditional brick and mortar stores. Over time, there has been a significant increase in the volume of electronic transactions. However, this growth has also drawn fraudsters who want to exploit every aspect of e-commerce, including payment methods, customer information, and goods. Criminals target businesses that transition or grow from brick and mortar to online platforms, claims. In order to defend and prevent e-commerce fraud, it is essential to identify it when it occurs. Online fraud detection techniques for online credit card transactions have made use of artificial intelligence. In addition to earlier methods, new techniques such as decision trees, K-Nearest Neighbors (KNN) fuzzy logic systems, Artificial Neural Networks (ANN), Bayesian networks, Hidden Markov Model, and Support Vector Machines (SVM) are provided. The results of the decision tree, SVM, and k-nearest neighbor algorithms are all moderately accurate. Out of all the algorithms, Fuzzy Logic has the lowest accuracy.