Risk assessment model training method and risk prediction method

• Main Authors: LOU ZONG, ZHANG WENTIAN, YAN DAQIANG
• Format: Patent
• Language: Chinese; English
• Published: 01.03.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The embodiment of the invention discloses a risk assessment model training method and a risk prediction method. The training method comprises the steps of obtaining sample data sets of a plurality of lessee, wherein each sample data set comprises a plurality of feature data representing operation information and risk information of the lessee and risk probability truth values; inputting a plurality of feature data of each sample data set into the risk assessment model for local training, performing prediction and judgment by using a loss function, and training to obtain a local optimal parameter of the risk assessment model corresponding to each sample; performing global training on all samples by using a cost function, and when the cost function is minimum, finishing training, obtaining global optimal parameters of the risk assessment model and outputting the trained risk assessment model; the parameters of the risk assessment model comprise the weight of each piece of feature data. The risk prediction metho