Shap-instance weighted and anchor explainable ai: enhancing xgboost for financial fraud detection

HIGHLIGHTS

• What: The study used a multiple regression model and identifies a strong and statistically significant correlation between various factors, including the size of the audit firm, auditor rotation, industry specialty, auditor market focus, auditor independence, audit report delay, and the identification of fraud through the restatement of financial statements. To bridge this gap, the authors propose the following research questions: RQ1: How can the authors enhance the performance of a model for fraud detection by relevant instance weighting? Unlike traditional methods that treat all instances equally, the authors propose the use of SHAP (SHapley Additive exPlanations) values . . .
  

   

  

  If you want to have access to all the content you need to log in!

  Thanks :)

  Username or Email

  Password

  Remember me

  Lost your password?

  If you don't have an account, you can create one here.