市场化取向改革中客户分档标准制定方法

发布时间：2018-05-13 14:07

本文选题：客户分档 + k-means　；参考：《中国烟草学报》2015年S1期

【摘要】：为贯彻落实2015年全国烟草工作会议精神,国家烟草专卖局扩大了卷烟营销市场化取向改革试点工作。其中,客户分档标准制定是卷烟营销工作的基础,其好坏直接影响到货源投放和新品培育工作。客户分档标准基于客户经营能力,可以根据进货量、进货金额等因素评分。根据客户评分,本文使用数据挖掘中的k-means聚类算法自动对其分类,并使用肘方法确定最佳分类数目。客户类别确定后,本文使用近邻分类kNN算法,加以适当调整,使之可根据客户商圈、业态等客户属性确定客户类别用于指导新入网客户的客户分类工作。本文提出的方法避免了依赖经验完成客户分档过程的主观性,提高了准确性,使得客户分档更加公平公正,同时也减少了客户分档的工作量。
[Abstract]:In order to carry out the spirit of the National Tobacco work Conference in 2015, the National Tobacco Monopoly Bureau expanded the pilot work on the market-oriented orientation reform of cigarette marketing. Among them, the establishment of customer classification standard is the basis of cigarette marketing, which directly affects the supply of goods and the cultivation of new products. Customer grading standard is based on customer management ability and can be graded according to quantity of purchase, amount of purchase and so on. According to the customer score, this paper uses the k-means clustering algorithm in data mining to classify it automatically, and uses elbow method to determine the best classification number. After customer classification is determined, this paper uses the nearest neighbor classification kNN algorithm to adjust it properly, so that the customer classification can be determined according to the customer's attributes, such as the customer business circle and business status, to guide the customer classification work of the new customers entering the network. The method proposed in this paper avoids the subjectivity of relying on experience to complete the customer filing process, improves the accuracy, makes the customer filing more fair and fair, and also reduces the workload of the customer filing.
【作者单位】：上海烟草集团有限责任公司贸易中心计算机信息部;
【分类号】：F721;F713.5

【参考文献】