Method for mining representative sequence pattern from Web click stream data

Abstract

The invention provides a method for mining a representative sequence pattern from Web click stream data, and relates to the technical field of sequence pattern mining. The method comprises the following steps: firstly, inputting a Web click stream sequence data set, a minimum support degree and a maximum coverage degree, traversing the data set once, and reserving all frequent sites which are not less than the minimum support degree as sequence generation seeds; adopting a gap expansion enumeration tree for each sequence generation seed, and combining a gap scanning pruning strategy and closure check to obtain all frequently closed super sequences of the seed; further selecting all representative sequences of the seed by adopting a local representative sequence screening technology; traversing all the sequences to generate seeds, and outputting a representative sequence of each seed to obtain all representative sequence modes of the Web click stream data. The method has the advantages that the representative sequence pattern can be used to effectively overcome the contradiction that the frequent sequence pattern is large in number and low in availability, and the availability of the result can be enhanced; and reference is provided for applications of online user behavior analysis, information recommendation, engine optimization and the like of the Web click stream.

Description

technical field [0001] The invention relates to the technical field of sequence mining, in particular to a method for mining representative sequence patterns from Web click stream data. Background technique [0002] Frequent sequences in sequence data are widely used in commercial fields such as Web click stream data analysis, customer shopping habit analysis, log data analysis, etc. Many users often buy "Coke" after buying "Potato Chips". Therefore, "Coke" can be recommended to customers who have purchased "Potato Chips" to increase the sales of the product. Sequential pattern mining can not only be applied to the commercial field, but has been widely used in many other fields, such as traffic travel pattern analysis, scientific experiment process analysis, natural disaster prediction analysis, disease drug diagnosis analysis, bioinformatics data analysis, etc. Among them, Web clickstream data analysis plays an important role in the fields of online user behavior analysis...

AI Technical Summary

Problems solved by technology

However, the existing sequential pattern mining methods have the following problems: On the one hand, most of the existing sequential pattern mining methods are completed based on two steps of "mining" and "selection", and there may be multiple scans during mining and selection. In the case of databases, it is difficult to deal with exponentially growing sequence data, which makes the calculation efficiency low; on the other hand, the existing sequence pattern mining methods produce a large number of mining results, and there are many redundant sequences and containment relationships sequence mode, which makes it impossible for users to understand, use and make decisions in a short time

Images