Project for Explore Data Science Academy
Once the dataset is imported it will need to be reviewed and cleaned . I have imported the cleaned data and ran analysis using PowerBi with screen shots provided.
After analysing the data is is evident that the company will need to flag claims that are filed fromeducation level : MD for major damaged vehicles as there is a great liklihood that that claim is fraudulant. The company should also request police reports for every claim.Implement FRAUD DETECTION models , Educate staff , Collaborate with law enforcement and continuously monitor and evaluate these strategies and adjust when needed . By doing this the company can improve its fraud detection and reduce losses associated with fraudulant claims
I built a Randon Forrest Regression Model to solution the issue this comapny is having with fradulent vehicle claims. The model MAE indicates on average the models predictions are off by 0.225units.