Study On Big Data Medical Insurance Audit In City F

In recent years,the scale of income and expenditure of medical insurance fund has been continuously increasing,and the scope of insurance has also been expanding.The contradiction between the ability and the needs of the medical insurance fund supervision is becoming increasingly prominent.Corroding medical insurance fund behavior occurs more frequently,and the techniques are more complex and more subtle,as decomposed hospitalization,false hospitalization,false prescription and items,and excessive prescription.Restricted by many factors,such as audit thinking,audit methods,auditors' ability,audit data and audit techniques,it is difficult to effectively meet the needs of the work by using traditional medical insurance auditing methods,and the auditing work is facing new challenges.Big data medical insurance audit is a great innovation in audit thinking,audit technology and audit methods.It can deeply discover the bug hidden in the massive data of medical insurance fund management,so as to promote the execution of medical insurance policy,deepen the reform of medical insurance system,strengthen the management of medical insurance fund,protect and improve people's livelihood.At present,big data medical insurance audit is still in the theoretical research stage,and there are few successful application cases both in domestic and overseas.By the use of literature research method,quantitative analysis method,and quantitative and qualitative analysis method,based on the past anemic heart disease data in City F,this article tries to construct K-means Cluster Analysis big data audit model and Naive Bayesian Classification Analysis big data audit model by combining audit immune system theory,DRGs(Diagnosis Related Groups)theory and data mining theory,applies these models to anemia heart disease in City F,and analyses the cases of heart disease patients corroding the medical insurance fund.In the case of unknown diagnosis and treatment characteristics,the former uses special algorithm to automatically analyze the data and divide them into several groups to generate significant characteristics and quantitative indicators of corroding medical insurance fund for anemic heart disease patients in City F.Combined with the previous audit results,the latter analyzes the characteristics of patients with corroding medical insurance fund,reckons the probability of patients with these characteristics of corroding medical insurance fund,and gets behavior characteristics probability table of anemic heart disease patients corroding medical insurance fund in City F.Using characteristic indicators and probability tables,auditors in City F can quickly select those patients with significant characteristics and high probability from the vast amount of medical data for highly suspected targets to check.Big data medical insurance audit finally expands audit thinking,enriches audit methods and improves audit quality.In addition,this paper further studies the safeguard measures of big data medical insurance audit in City F from the aspects of audit thinking,audit system,audit method and personnel training,and puts forward the following measures: changing audit thinking,establishing big data audit system,strengthening personnel training,establishing big data analysis platform,and opening up with data-first audit method The development of big data audit projects has provided new ideas and solutions for medical insurance audit in City F.