AprioriAll算法的基本思路

1) 排序阶段 利用客户标识customer 2id作为主关键字以及事务发生的时间transaction 2 time作为次关键字对数据库D排序,该步骤将原始的事务数据库转换成客户序列的数据库.

2) 发现频繁项集阶段 利用关联规则挖掘算法找出所有的频繁项目集.

3) 转换阶段 在已经转换的客户序列中,每一个事务被包含于该事物中的所大项目集来替换,如果一个序列不包含任何大项目集,则在已经转换的序列中不应该保留这项事务.

4) 序列阶段 利用核心算法找出所有的序列模式.

-Sequential pattern mining from the sequence found in the database as a sequence of frequent pattern, it is a kind of important data mining issues, has a very wide application, be used in customer buying behavior, including the analysis of network access mode of analysis, the scientific experiments Analysis, the early diagnosis of disease, natural disasters forecast, DNA sequences deciphered, and so on. The efficiency. In this paper, I was in the sequence pattern mining one of two algorithms to study, namely: Armorial and GSP algorithm. First on　the　sequence patterns of some basic concepts and principles. And demonstrate through concrete examples of the implementation of the algorithm, then reached into the grasp of understanding. Used vc again based on the programming language and Access database to achieve the end result of running the analysis and synthesis.