Data Science has been successfully applied in various areas of fraud detection and prevention, such as credit card fraud, tax fraud, and wire transfer fraud. However, there is insufficient research on the use of data mining in fraud related to internal control. We analyze the data set on work time claims for projects using two techniques: the decision tree of automatic chi-square interaction detection (CHAID) and link analysis. The results indicate that the following characteristics of putative work time claims were the most significant: Client industry, consultant background and expertise, and cost of consulting services. Our research contributes to the field of data-driven internal control with the goal of preventing fraudulent work time claims in project-based organizations.