Multi-account relevance method and device, and electronic equipment

Abstract

The invention discloses a multi-account relevance method and device, and electronic equipment. The multi-account relevance method comprises the following steps: obtaining the URL (Uniform Resource Locator) information of a user network behavior; according to a network entry rule, regularly mining account information from the URL information; filtering the account information, and ejecting dirty data; on the basis of the URL information and a Refferer message, generating multiple pieces of Session for the filtered account information at set time intervals; extracting the account information under the same Session to generate an account information group with a relevance relationship; according to the relevance relationship, calculating the support degree and the confidence coefficient of each account information group; and on the basis of the support degree and the confidence coefficient, carrying out graph operation on the account information group to generate an image of the relevance relationship. The multi-account relevance method fully mines the Internet access behavior information of a user and improves the classification accuracy and integrity of the user account information.

Description

technical field [0001] The present invention relates to the field of data mining, in particular to a multi-account association method, device and electronic equipment. Background technique [0002] With the continuous expansion of the field of big data, the data of Internet users is increasing day by day. Internet user data includes user information such as user age, gender, occupation, etc. This information penetrates into major Internet fields such as social networks and e-commerce, and has great economic value. , Building user portraits, etc., play a decisive role. [0003] The precondition for obtaining user data is user mining and identification. Existing technologies are usually based on the same IP (Internet Protocol) or coarse-grained association rules such as IP and User Agent (UserAgent). The identity information (Identity, ID) of the cookie data is forcibly associated with the application account information. [0004] However, due to the complexity and diversit...

AI Technical Summary

Problems solved by technology

[0004] However, due to the complexity and diversity of network interconnection, the same user often has multiple website accounts or application program (Application, App) accounts; and there is a large amount of dirty data in the ID information and application program account information of the cookie data; An IP or IP and UserAgent may be used by different Internet users, resulting in inaccurate association rules, and thus cannot accurately associate and classify all account information data of the same user

Images