Fraud Detection in Workers Compensation Insurance: A Random Forest Approach using Public Enterprise Data

Jongchan Kim¹, Changyong Lee²

^1,2School of Management Engineering, Ulsan National Institute of Science and Technology, Republic of Korea

Abstract

Insurance fraud detection plays an important role in protecting the profitability of insurers and the insured people. The prevailing approach to detect insurance fraud used variables created by the opinions of domain experts and applied data mining and machine learning techniques. However, prior research is subject to three limitations, summarized as a dearth of research on public insurance, financial variables, and algorithms that can estimate the extent of the fraud. Therefore, we propose a fraud detection model by providing a concept of estimating the extent of public insurance fraud with workers’ compensation insurance premium. This research plans to use data obtained from several public enterprises, and by applying random forest model with multiple financial variables, the workers’ compensation insurance premium will be estimated. The quantitative outcome to be provided by the proposed approach enables decision makers in public enterprises and the government to offer finer policies for the citizens.

Keywords

Author's Profile

Author profile can be generated and linked through our partners World Book of Researchers. To include your profile online Click Here. After it is approved, please email to edlib @ asdf.res.in to create a link with all the papers.

• Jongchan Kim : Profile
• Changyong Lee : Profile

e-AID

Cite this Article as Follows

Jongchan Kim, Changyong Lee. Fraud Detection in Workers Compensation Insurance: A Random Forest Approach using Public Enterprise Data. International Conference on Business, Finance and Economics (2017): 09. Print.