基于极限学习机的汽车保险欺诈识别模型构建与研究

Construction and Study of Auto Insurance Fraud Identification Model Based on Extreme Learning Machine

随着保险业的发展,保险欺诈也在全球范围内蔓延,尤其在汽车保险领域。因此,从极限学习机的理论出发,对基于极限学习机的汽车保险欺诈模型进行研究,引入广义线性模型,提出了一种广义线性模型—极限学习机(GLM-ELM)汽车保险欺诈识别模型。首先进行汽车保险欺诈数据的筛选与处理,然后将广义线性模型用于参数估计和拟合数据分布,从而满足模型对数据分布的要求,最后将拟合分布后的数据输入到GLM-ELM汽车保险欺诈识别模型中,进行实证分析并得出结论。结果表明:相对于传统的模型而言,基于GLM-ELM的汽车保险欺诈识别模型能够更好地识别索赔数据中的欺诈信息。

With the development of insurance industry, insurance frauds occur all over the world, especially in auto insurance field . Therefore, a Generalized Linear Model was introduced into study of the auto insurance fraud model based on Extreme Learning Machine. We propose an auto insurance fraud detection model based on the Generalized Linear Model-Extreme Learning Machine （GLM-ELM）. First , we filter and process the data, and then the Generalized Linear Models was used to estimate the parameters and fit the data distribution to meet the requirements for data distribution . Finally we input the data with the fitted distribution to the GLM-ELM fraud detection model and draw the conclusions . The results show that ： Compared with the traditional model, the auto insurance fraud detection model based GLM-ELM can better identify the fraud information of the claims data.