Characteristic index screening method, satisfaction prediction model construction method and prediction method

• Main Authors: WANG GUANGSHAN, WANG ZONGHUI, CHEN JIKUO, GU LONG
• Format: Patent
• Language: Chinese; English
• Published: 15.04.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention relates to a feature index screening method, which comprises the following steps: obtaining a user data set, user data in the user data set comprising at least two feature indexes; at least two different pre-screening models are constructed based on the same n characteristic indexes, the at least two different pre-screening models form a pre-screening model group, and n is a natural number greater than 0; and when it is judged that the performance indexes of the pre-screening model group meet requirements, taking the n feature indexes as the screened feature indexes. The invention further relates to a satisfaction prediction model construction method and a satisfaction prediction method. The degree of satisfaction prediction model constructed by the method and the device is more universal. 本申请涉及一种特征指标筛选方法，包括：获取用户数据集，其中，所述用户数据集中的用户数据包括至少两种特征指标；基于相同的n种特征指标构建不同的至少两个预筛选模型，所述不同的至少两个预筛选模型形成预筛选模型组，其中，所述n为大于0的自然数；判断所述预筛选模型组的性能指标符合要求时，将所述n种特征指标作为所筛选出的特征指标。还涉及满意度预测模型构建方法及满意度预测方法。本申请所构建的满意度预测模型更具有通用性。