Stack Path Identification and Encryption informed by Machine Learning as a spoofing defense mechanism in E - commerce

Abstract

Spoofing attacks are a constant nag in the information world, so many methodologies have been developed to reduce on its effects but they have not been satisfactory. The kind of impact that this attacks have on Electronic Payment Systems is detrimental to the economic world given that this systems are viewed as performance enhancers on payments. A lot of resources are consumed thus giving rise to a situation that deserves undivided attention and should be researched on. The answer to the question does spoofing attacks affect the adoption of electronic payment systems is given.This research focused on a hybrid solution oftwo methodologies,a combination of StackPi and Encryption as spoofing defense mechanisms. Data was collected using a standardized self-administered questionnaire with a sample study where a sample of 10 technical staff experts and 4 management staff information technology experts participated. The research looked into details of types of defense mechanisms that have been implemented in the past to curb spoofing attacks, their limitations and how combining these techniques lead to better results. Machine learning was used and found to have resilient capabilities for its ability to use an already existing data set to learn the behavior of a spoofed data their Internet Protocol addresses and Path identification markings helped in updating of the filter table and trace back potential of maligned sources.Evaluation results showed that a synergy between machine learning filtering methods, and encryption provides an optimum defense mechanism against spoofing attacks.The results of the study can be of significant use in defense against spoofing attacks on organizations ’information systems.