基于数据过滤Apriori算法的图书推荐模型

A book recommendation model based on data filtering Apriori algorithm

可以从读者的图书借阅记录中挖掘有价值的数据,识别读书偏好,提供个性化的图书借阅推荐服务。Apriori算法存在单一用户的单一借阅记录在整体数据集中变成离群点,导致分析时间和内存开销显著增加的问题。通过设定置信度、支持度和过滤度的阈值,对原数据集进行过滤;再使用Apriori算法对新的数据集进行关联规则分析。带有数据过滤的关联规则算法在图书借阅记录数据量无论大和小的情况下,分析时间更短,内存开销更小,强关联规则更强。

It can mine valuable data from readers’book borrowing records,identify reading preferences,and then provide personalized book borrowing recommendation service.Apriori algorithm has the problem that the single borrowing record of a single user becomes an outlier in the whole dataset,which leads to the significant increase of analysis time and memory cost.By setting the threshold of confidence,support and filtering,the original data set is filtered,and then the Apriori algorithm is used to analyze the association rules of the new data set.The association rule algorithm with data filtering has shorter analysis time,less memory cost and stronger association rules in the case of large amount and small amount of data.