一种可视化的加权平均信息融合方法
A Visibility Graph Information Fusion Method with Weighted Average

针对传统加权信息融合方法不能有效降低信息损失、快速做出合理判断的问题,提出了一种可视化的加权平均信息融合方法。首先,将原始信息数据按数值大小进行降序排列,生成一个有序序列,再采用可视图转换法将该有序序列的数据转化到网络拓扑结构中,其中数据和网络节点一一对应;然后,利用支持度函数对网络中的节点关系进行衡量,从而得到原始数据的权重;最后,将原始数据进行加权融合,得到最终结果。与经典的融合方法相比,新方法从一个更客观的角度对数据权重进行处理,准确度更高,对数据融合及决策问题的处理更有效。实验结果表明:新方法具有更好地实际可操作性和合理性;更重要的是,新方法降低了融合过程中的信息损失,几乎不会产生与直觉相悖的融合结果,大大提高了人们在决策分析处理中的有效性。
A visibility graph fusion method with weighted average is proposed to solve the problem that the classical weighted averaging fusion methods cannot effectively reduce the information loss and make reasonable judgments quickly. First, argument values are ordered in a descending order to form a time series. After that, a network is constructed based on the visibility graph, and the time series is converted into a graph. The data and network nodes correspond one to one. Then a support function is proposed to measure the relationship between the nodes, and the weights of original data are obtained. Finally, the original data are weighted and fused to get final results. Compared with the classical methods, the new method deals with the weights from a more objective perspective, and it is more accurate and effective when dealing with decision making problems. An example of produced water management is presented to show that the proposed method is reasonable and practical. Moreover, this method reduces the information loss in the fusion process, and hardly produces a counter？？intuitive result. It greatly improves the effectiveness of decision analysis