改进RGM的用户情景状态序列信息预测方法

Sequence information prediction method of user contest states based on improved RGM

用户情景状态序列信息缺失补齐是多数据源数据融合研究中数据预处理的关键问题之一.针对用户的情景数据序列中存在信息缺失的问题,在滚动灰色预测模型(RGM,rolling grey model)的基础上提出一种改进方法.该方法对已知用户情景状态原始数据序列进行幂指变换,通过变换后的数据进行滚动灰色预测并将预测出的数据进行幂指逆变换,从而得出情景状态缺失数据的预测值.实验结果表明,该方法在很大程度上提高了预测数据的精确性.

The lack of sequence information of user context states is one of the key problems to data pretreatment in multivariate source fusion research. In view of the missing user information data,it proposed an improved method based on the Rolling Grey Model. The method made the original data sequence of the known users context states go through exponential transform. The transformed data was processed in the rolling grey prediction and the predicted data was made to undergo the exponential inverse transform. Finally the prediction results of context states were obtained. Experimental results show that the method can greatly improve the accu- racy of the prediction data.