实验二：分类挖掘等.docVIP

SQL SERVER 数据挖掘实习教程 二、决策树、贝叶斯、聚类挖掘综合实习 Targeted Mailing The marketing department of Adventure Works is interested in increasing sales by targeting specific customers for a mailing campaign. By investigating the attributes of known customers, they want to discover some kind of pattern that can be applied to potential customers, which can then be used to predict who is more likely to purchase a product from Adventure Works. Additionally, the marketing department wants to find any logical groupings of customers already in their database. For example, a grouping may contain customers with similar buying patterns and demographics. Adventure Works contains a list of past customers and a list of potential customers. Upon completion of this task, the marketing department will have the following: A set of mining models that will be able to suggest the most likely customers from a list of potential customers A clustering of their current customers 有针对性的邮件 历险公司的市场部对增加销售目标感兴趣，目标是针对特定的客户进行邮寄广告活动。通过对已知客户的属性进行调查，他们希望发现一些可以应用于潜在客户的模式，这些模式可以用来预测谁更有可能从冒险活动中购买产品。 另外，市场部门希望找到已经在数据库中的客户的任何逻辑组合。例如，一个分组可能包含具有类似购买模式和人口统计特征的客户。 冒险作品包含了过去的客户列表和潜在客户列表。 完成这项任务后，市场部将会有以下的工作: ?一组挖掘模型,将能够建议最可能的客户从潜在客户列表中 ?当前客户的聚类 In order to complete the scenario, you will use the Microsoft Na?ve Bayes, Microsoft Decision Trees, and Microsoft Clustering algorithms. The scenario consists of five tasks: Create the mining model structure. Create the mining models. Explore the mining models. Test the accuracy of the mining models. Create predictions from the mining models. 为了完成这个场景，您将使用Microsoft朴素贝叶斯、微软决策树和微软集群算法。该方案包括五个任务: ?创建挖掘模型结构。 ?创建挖掘模型。 ?探索挖掘模型。 ?测试挖掘模型的准确性。 ?创建挖掘模型的预测。 Create a Targeted Mailing Mining Model Structure Using the Wizard The first step is to use the Mining Model Wizard to create a new mining structure. The Mining Model Wizard also creates an initial mining model based on the Microsoft Decision Trees algorithm. 使用向导创建一个目标邮件挖掘模型结构 第一步是使用挖掘模型向导创建新的挖掘结构。挖掘模型向导还创建了基于Microsoft Decision Trees算法的初