Feature extraction and individuation recommendation method and system based on user behaviors

Abstract

The embodiment of the invention provides a feature extraction method based on user behaviors. The method comprises the steps that click stream data of access to the webpage of users are obtained; according to the click stream data, the relevancy of a path through which the current user gets access to the webpage and paths through which other users get access to the webpage is calculated; X former users with the highest relevancy to the path through which the current user gets access to the webpage are extracted, wherein X is a positive integer; the comprehensive weight is configured according to a preset label of the webpage to which the X former users get access; the preset label and the comprehensive weight are adopted for calculating the relevancy between the current user and the X former users. Based on the click stream data, a digraph model, with the weight, of the click path of access to the webpage of the users is constructed, the relevancy calculation of the users is converted into the similarity calculation of a digraph with the weight at first, a webpage label bank is introduced, calculation of the relevancy of the webpage label content is fused, the click habit and the individual behavioral preference of the users are mined, and therefore the user cluster accuracy and efficiency are improved.

Description

technical field [0001] The embodiment of the present application relates to the technical field of data processing, in particular to a feature extraction method based on user behavior, a system for feature extraction based on user behavior, a personalized recommendation method based on user behavior, and a user behavior-based Personalized recommendation system. Background technique [0002] Internet The rapid development of the Internet has brought people into the era of information society and network economy, and has had a profound impact on the development of enterprises and personal life. At the same time, the excessive information makes it impossible for people to efficiently obtain the parts they need, and the efficiency of information use decreases instead. [0003] Taking e-commerce (Electronic Commerce) as an example, e-commerce is a new type of online shopping, online transactions between merchants and online electronic payment based on the browser / server applicat...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, user clustering is mostly based on user access paths or user search keywords based on the distance between webpage click sequences. User access is generally multiple and intermittent, and the click paths of each visit cannot be exactly the same. It cannot comprehensively measure the differences brought about by users' multiple visits, and the effect of user clustering is poor, resulting in low service efficiency

Images