Abnormal Detection Method of Pension Meal Service Data Based on E-ARLL Algorithm
China is entering an aging society.In order to ensure the three meals a day for the elderly,the government purchases pension meal services in various places.The false services,embezzlement,falsely use and other problems in the service process threaten the property security of government and the elderly,so this paper proposes E-ARLL algorithm to detect abnormal data.This method uses Pearson correlation coefficient and ANOVA to divide the original dataset into the feature training set and the feature verification set,and then the feature training set is input into the E-ARLL algorithm model.Based on the thinking of Ensemble Method,the suitable algorithm is selected for abnormal detection according to the linear relationship of the divided dataset.The experimental results show that the proposed method shows good performance on the abnormal detection of pension meal service data,and the final abnormal data identification rate is 99.4%.It brings new verification methods to the credibility of government purchasing services,which has profound significance.
abnormal detectionintegrated algorithmpension service
国家科技期刊平台
NSTL
万方数据
维普
