The Effectiveness of User Input for Fraud Detection in the Supply Chain Management Field
This research paper investigates the effectiveness of user input to assist with fraud detection in the supply chain management industry. To research this, a software application is created to mimic a supply chain management system and an internal component is engineered to perform the fraud analysis of transactions recorded in a database. The analysis module makes use of the Logistic Regression and Decision Tree Machine Learning algorithms. The test results show that user input does assist with detecting fraudulent transactions if there is a large number of transactions present for analysis and training the system. The test results indicate that the better machine learning algorithm to use is Decision Trees. Based off the test results and training data, the use of user input does assist with detecting fraudulent activity in supply chain management if there is enough training and testing transaction data to analyze.
|The Effectiveness of User Input for Fraud Detection in the Supply Chain Management Field
|CC BY 4.0 (Attribution)
|Masters Culminating Experience
|November 18, 2022
|January 03, 2023