An account association method for cross-UGC website platform

Abstract

The invention discloses a method for associating accounts belonging to the same entity user on multiple UGC websites. The basic principle is to extract features from text content generated by UGC website accounts to associate users belonging to the same entity on different UGC website platforms multiple accounts. The method consists of data acquisition, data preprocessing, feature extraction and layer-by-layer filtering. The data acquisition part collects the text content generated by the user account of the target UGC website. The data preprocessing part preprocesses the text content. The feature extraction part extracts gender features, age features, geographic location activity features and writing style features from the text content. The layer-by-layer filtering part filters out accounts irrelevant to a given user account layer-by-layer by gender, age, geographic location activity, and writing style features in turn. The invention can effectively solve the problem that accounts of the same entity user on different UGC websites are irrelevant, and has high practical value.

Description

technical field [0001] The invention belongs to the field of Internet account analysis, and more specifically relates to a method for associating multiple accounts for cross-UGC website platforms for users belonging to the same entity. Background technique [0002] UGC (User Generated Content) websites mainly refer to online social networking sites and online forums based on user-generated content. At present, the famous UGC websites include Facebook, Twitter, Google+, Renren.com, Kaixin.com, Douban, Sina Weibo, Tencent Weibo, etc., have become the mainstream network platforms for sharing information. However, various UGC websites provide different types of services to netizens. For example, Facebook, Google+, and Renren mainly provide social services for users. Twitter, Sina Weibo, and Tencent Weibo allow users to share their latest developments and ideas. Taobao and JD.com are mainly used for e-commerce. Therefore, the information we obtain will be different in different...

AI Technical Summary

Problems solved by technology

[0007] The purpose of the present invention is: aiming at the deficiencies of the prior art, through improvements in account modeling and account association methods, a new user account association method for cross-UGC website platforms is proposed, which solves the problem of traditional account association methods User characteristics are unstable, the error is relatively large, and user characteristics cannot be fully and effectively used, thereby improving the accuracy and robustness of user account association across UGC website platforms

Images