Performance Evaluation Of Social Endowment Insurance System Based On RS-SVM

The evaluation of the operation of social endowment insurance system is the basis of social security system reform and innovation.This paper applies rough set and support vector machine to the performance evaluation of social endowment insurance system in China.The combination of the two can not only exert their respective advantages,but also have strong complementarity,which can better solve many shortcomings in the performance evaluation of the social endowment insurance system.The main contents of this paper are as follows:(1)This paper evaluates the operation status of China's social endowment insurance system from three aspects:fund revenue and expenditure level,fund management,endowment insurance and economic and social aspects,and refines each level,and selects 14 performance evaluation indicators of social endowment insurance system.(2)On the basis of introducing the basic theory of rough set and support vector machine,the advantages of combining the two methods are analyzed.Therefore,RS-SVM is proposed to be applied to the performance evaluation model of social endowment insurance system.Rough set is used to calculate the relative importance and weight of each index to the overall decision-making,and the index with zero importance is deleted to establish a new decision-making table on the premise of ensuring no difference.The number of samples in the new decision-making table is divided into five groups on average,four groups are randomly selected as training sets,and the rest as test sets.The grid method is used to find the optimal parameters on the training set,and then the prediction model is established.Finally,the model with the best fitting effect of the test set is chosen as the optimal model.(3)In the"China Statistical Yearbook",we collect the relevant data of the performance evaluation indicators of social endowment insurance system,that is,conditional attribute values;we collect the attribute values of decision-making using expert scoring method.Firstly,rough set is used to calculate the relative importance and weight of each evaluation index,and redundant index is deleted to simplify the decision table.Then,support vector machine is used to establish the performance evaluation prediction model of social endowment insurance system.Five fold cross-validation method is used to test the validity of the model.It is concluded that the impact of rural endowment insurance coverage,urban registered unemployment rate and the contribution rate of rural insured personnel on the performance evaluation of social endowment insurance system can be neglected,and the performance evaluation of social endowment insurance system is constructed and the model of the performance evaluation of the social endowment insurance system is established:(?)...