论文部分内容阅读
以客户关系管理(custom er re lationsh ip m anage-m en t,CRM)的数学模型为背景,研究了如何用Hop fie ld神经网络构建一类M arkov链表述的CRM的客户分类分析和评价的计算模型。从一种不同于现CRM的对客户关系评价的思路入手,用M arkov链建模。分析该类马氏链建立的CRM数学模型的特点(无限次交易),分析连续Hop fie ld神经网络计算的内在特点。从矩阵结构和求逆的角度,发现这两个不同概念模型的数学模型具有相同的特点。研究结果将该类马氏链的CRM模型计算同Hop fie ld神经网络计算关联起来。这表明可以用连续Hop fie ld神经网络计算该类M arkov链的CRM模型。
Based on the mathematic model of CRM (Customer Relationship Management), this paper studies how to classify and evaluate CRM clients based on Hopfield neural network to construct a M arkov chain representation Calculate the model. Starting with a different approach to customer relationship evaluation than present CRM, M arkov chain modeling. The characteristics of the CRM mathematical model established by this kind of Markov chain (infinite trading) are analyzed, and the inherent characteristics of the continuous Hopfield neural network are analyzed. From the perspective of matrix structure and inversion, we find that the mathematical models of these two different conceptual models have the same characteristics. The research results relate the CRM model calculation of this kind of Markov chain with the Hopfield neural network. This shows that the CRM model for this type of M arkov chain can be calculated using a continuous Hopfield neural network.