决策演化集的伯努利移位

Bernoulli shift of decision evolution set

决策演化集是一种新的用来解决决策规则在时间序列上变化问题的方法。决策演化矩阵是其中一个重要概念。通过决策演化矩阵可以直观了解到各个条件属性在单个时间点以及连续时间序列上的表现。但目前的体系中虽然对决策演化矩阵进行了定义,但还缺乏数值量化,文章引入伯努利移位来对决策演化矩阵进行数值量化,并通过实例演示这些量化后的数值在预测中所起的作用。

Decision evolution set is a new method to deal with evolution problem of decision rules changing in time series.Decision evolution matrix is an important concepts of decision evolution set.Through the decision evolution matrix,we can intuitively understand the performance of each conditional attribute at a single time point and a continuous time series.However,the decision evolution matrix is defined in the current system,there is still a lack of numerical quantization.In this paper,Bernoulli shift is introduced to quantify the decision evolution matrix,and an example is given to demonstrate the role of these quantized values in prediction.