基于SCNBMF的智慧城市不完备数据的处理系统及方法

• Format: Patent
• Language: Chinese
• Published: 28.06.2024
• 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

本发明公开了一种基于SCNBMF的智慧城市不完备数据的处理系统及方法，不完备数据处理系统将训练样本建立一个区域－特征矩阵，利用自洽性的预填充模型对训练样本中包含的缺失值进行估计和填充，再借助基于负二项分布的填充模型挖掘隐性特征，在前一阶段所得的预填充结果的基础上，进一步改善填充结果。应用本发明方法能够准确估计出智慧城市不完备数据的真实数值，并能有效地还原智慧城市不完备数据中的隐性结构。 The invention discloses a smart city incomplete data processing system and method based on SCNBMF, and the system builds a region-feature matrix for a training sample, carries out the estimation and filling of missing values in the training sample through a self-consistent pre-filling model. And mining recessive features by means of a filling model based on negative binomial distribution, and further improving the filling result on the basis of the pre-filling result obtained in the previous stage. By applying the method, the real numerical value of the smart city incomplete data can be accurately estimated, and the hidden structure in the smart city incomplete data can be effectively restored.