基于改进Apriori算法的客户需求数据分析方法

Customer Demand Data Analysis Based on Improved Apriori Algorithm

为了更好的利用与产品参数方面的客户需求数据,提出了一种基于布尔矩阵改进的Apriori算法对客户需求数据信息进行分析。首先,针对Apriori算法在每次由低维度连接生成高维度的候选频繁项集时都需要扫描整个数据库非常耗时的缺陷,利用布尔矩阵对其进行改进,把客户需求数据映射成布尔矩阵;其次,采用迭代和剪枝的方式,利用改进后的算法对客户需求数据进行分析,计算出满足设定支持度的最高维度的频繁项集,挖掘出客户需求信息之间的不确定性联系,为设计制造出满足客户需求的产品提供参考。通过分析,改进后的算法在计算的时间复杂度和空间复杂度方面更优;最后以某企业针对冰箱产品开展的客户需求调查结果为例,说明该方法的具体实施过程。

In order to better utilize the customer demand data of product parameters,a new Apriori algorithm based on Boolean matrix is proposed to analyze customer demand data. First,because the process of scanning the entire database is time-consuming when Apriori algorithm generates candidate frequent item-sets of high dimensions from low dimensions,Boolean matrix is established to improve this process and build mapping relationship from customer demand to Boolean matrix.Using Boolean matrix to improve the Apriori algorithm. Mapping the customer demand data into Boolean matrix;Second,the improved algorithm with method of iteration and pruning is used to analyze the customer demand data and the frequent itemsets with highest dimension are calculated,which satisfy the setting support degree. Digging out the uncertainty relationship among the customer demand information to provide reference for the design and manufacture of products which meet customer demand. Through the analysis,the improved algorithm is better in computing time complexity and space complexity;Finally,an example of the results of customer demand survey in a certain enterprise is given to explain the implementation process of this method.