Research On The Identification And Prevention And Control Of Claims Risk Of Accidental Injury Insurance

In recent years,with the improvement of people’s living standards,the insurance industry has shown a spurt of development.People’s insurance awareness is also improving.Based on the frequent occurrence of accidents,accident insurance can provide a guarantee for families and avoid family members’ "streaking".Therefore,more and more families attach great importance to the purchase of accident insurance（accidental injury insurance for short,the same below）.However,insurance and insurance fraud go hand in hand.Since the emergence of insurance,insurance fraud and insurance fraud have emerged in an endless stream,especially for the payment of huge insurance compensation insurance such as accident insurance.Insurance fraud cases of accident insurance have occurred repeatedly in recent years.Once insurance fraud is successful,it will cause great economic losses to insurance enterprises.It not only increases the insurance cost of the honest insured and disturbs the market economic order,but also the appearance of fraud will lead to casualties.Insurance fraud is an important factor threatening social security and economic stability.In response to this phenomenon,I have read and deeply studied the relevant literature and books at home and abroad.With Based on a careful study,summary and analysis,I decided to deeply analyze the claim risk of accidental injury insurance,and give reasonable suggestions through the conclusion,hoping to make some modest contributions to the healthy development of the insurance industry.Based on the theory of risk assessment,this paper uses data mining theory and criminal science and technology principles and methods to study the risk assessment in the accident insurance claims process through a combination of literature reference,mathematical calculation model application and software tool analysis.The contents of this paper include:the definition of the concept of accident insurance fraud,the methods of identifying insurance fraud at home and abroad,the application of criminal science and technology and data mining technology in the identification of accident insurance,as well as the problems of information asymmetry,moral risk,adverse selection,technical risk and human life value,etc.The current situation,causes and classification of insurance fraud of accident insurance in China.Identify fraud factors and use data mining theory to analyze the network relationship,score and grade the fraud factors,set up key items for the key factors,and establish the identification system.In view of the advantages and disadvantages methods and applicable conditions of fraud identification and the current accident insurance investigation in China,the method of combining logistic regression with K-means clustering analysis of data mining and decision tree classification processing model is selected as the empirical analysis method in this paper.On this basis,an empirical analysis of the claims settlement of a large life insurance company in my country,and uses the confusion matrix to draw the ROC curve,AUC curve,cumulative ROI diagram and cumulative lifting chart to evaluate the accuracy and rationality of the identification and prediction system.Put forward the countermeasures to prevent accidental insurance fraud and look forward to the sound development of my country’s insurance market.