基于跨区域差异残缺数据分析的数据补齐方法

Method of Data is Lacking Based on Inter-Regional Differences Incomplete Data Analysis

提出了基于支持向量机以及量化容差关系的数据补齐模型,针对不同区域数据塑造差异残缺数据的支持向量机预测模型,实现对跨区域差异残缺数据的预测,采用RBF神经网络修补残缺数据,通过基于量化容差关系的残缺数据补齐方法对残缺数据进行深度补齐,实现对跨区域差异残缺数据的进一步优化。实验结果显示了该种方法进行的跨区域差异残缺数据补齐效果优于传统方法。该种方法具有较高的补齐准确率,可获得满意的修补效果。

Is proposed based on support vector machine （SVM） and quantitative data filling model of tolerance relation, to shape differences of different regions of the support vector machine forecasting model of incomplete data, to achieve cross-regional differences of incomplete data prediction, RBF neural network was adopted to repair the damaged data, through quantitative tolerance relation based on incomplete data completion method on the depth of incomplete data is lacking, achieve cross-regional differences to optimize the incomplete data. Experimental result shows the approach across the regional difference of incomplete data supplement the effect is better than the traditional methods, this method has higher filling accuracy, a satisfactory repair effect can be obtained.