设为首页|加入收藏|网站地图|RSS
请选择搜索分类 新闻 留学 专题

数据挖掘在影院客户关系管理系统中的应用研究

发布时间:2011-12-17 16:47 来源:工大在线字号:T|T

  数据挖掘在影院客户关系管理系统中的应用研究
  
  摘    要
  
  客户关系管理(CRM)系统的运用,可以有效地帮助企业开展营销活动,全面迎合用户的需求,为其提供合适的产品与服务。然而当企业发展了多年,拥有了大量的客户数据,如何有效地利用这些数据,分析出对于企业有用的知识,进而采取恰当的经营策略来改善客户关系实现利润最大化,是每家企业所面临的最大问题。数据挖掘正是从大量的历史数据中发现隐含的、有潜在应用价值并最终能形成被人理解的知识过程。数据挖掘和CRM结合,通过数据挖掘优化CRM流程,运用数据挖掘得到的知识可以大大提高促销的效果,帮助销售人员更准确地定位经营活动,并使活动紧密结合现有客户和潜在客户的需求、愿望和状态,提高客户的忠诚度,实现经营最优化。
  
  本研究通过分析某影院的历史销售数据,提出CRM系统下数据挖掘的两种应用:运用数据挖掘中的K-Means聚类算法,对客户的消费行为进行挖掘得出消费群体的聚类规则,依据分析得出的观众电影消费习惯,可以有针对性地开展促销活动;通过分类回归树算法对影片的销售进行分析,实现对影票销售的市场预测,为影院的排片提供了决策支持。
  
  使用聚类算法进行客户划分,得到8类特征明显的观众群体,有效地优化了影院的经营策略。CART算法对离散值变量进行分类,对连续值变量进行回归,得到一个影票销售预测的分类回归树模型。该模型训练数据的值与验证数据的值基本接近,预测效果能满足提供影院排片决策支持的需求。
  
  本研究对实施影院客户关系管理系统有一定的指导作用。
  
  关键词:数据挖掘,客户关系管理(CRM),K-Means算法,分类回归树(CART)
  
  Abstract
  
  With the wide application of Customer Relationship Management(CRM) system, enterprise could unfold activities to meet the demand of their customers and provide them with the right products and services. After the running of CRM systems for some time, however, the enterprise may have enormous amount of customer related data. This poses the question for decision makers:How to convert large volume of data into valuable knowledge? How to use the knowledge to guide enterprise activities? Those are big challenges. Data mining is the process of discovering connotative, previously unknown, potentially useful and understandable information from large datasets. Data mining technology combine with CRM system can optimize its procedure. When make use of the knowledge mined by this system,it would improve the sales promotion greatly,help the manager force to orient the business activities more accurately, and make the activity accord with the demand , hope and state of existing customer and potential customer closely, moreover improve the customer's loyalty,realize that management is optimized most.
  
  This research discusses tow kinds of data mining application under the analysis of a cinema historical running data. Using the K-Means cluster Algorithm to cluster the groups of customer's consumptive behavior and get audience's film consumption habit .Then can develop the activity of promoting pointedly;Having analyzed the sales of the films through Classification and Regression Tree Algorithm, can forecast the ticket sales of the cinema ,and offer decisions support for the arranging of the cinema showing too at the same time.
  
  Using cluster's algorithm to divide the customer, eight characteristic of consuming groups got is obvious. So it can effectively optimize the cinema's working strategy. Classification and Regression Tree arithmetic classify the dispersed value variables, regress the continuous value variable,and return a model of CART for tickets sales forecast. The training data value is close to the confirmation data value of this model, it can meet the demands for decision support to arrange cinema showings.
  
  The research of this text has certain guidance functions in implementing Customer Relationship Management System of the cinema.
  
  Key words:Data Mining, Customer Relationship, Management(CRM), K-Means Algorithm, Classification and Regression Tree(CART)

图库

更多

网友评论

排行

点击 评论 关注