Search International and National Patent Collections

1. (WO2017016059) TEXT MINING-BASED ATTRIBUTE ANALYSIS METHOD FOR INTERNET MEDIA USERS

Pub. No.:    WO/2017/016059    International Application No.:    PCT/CN2015/090747
Publication Date: Fri Feb 03 00:59:59 CET 2017 International Filing Date: Sat Sep 26 01:59:59 CEST 2015
IPC: G06F 17/30
Applicants: CHENGDU YUNDUI MOBILE INFORMATION TECHNOLOGY CO., LTD
成都云堆移动信息技术有限公司
Inventors: ZHANG, Guo hong
张国鸿
TAN, Xingjun
谭兴军
Title: TEXT MINING-BASED ATTRIBUTE ANALYSIS METHOD FOR INTERNET MEDIA USERS
Abstract:
Disclosed in the present invention is a text mining-based attribute analysis method for Internet media users. The method comprises the following steps: (1) sequentially establishing a label main corpus and a feature corpus, and updating and maintaining the label main corpus and the feature corpus respectively; (2) and extracting all history article samples of Internet users, and removing videos, audios and pictures in the samples. The present invention can form attributes of browsed sample articles for each of the Internet users, and accurately determine weights of interest categories through analysis, so as to deeply identify, analyze and mine the user attributes of the users, that is, in addition to the capability of analyzing and mining the basic attributes of the users, the application range of the identification of the user attributes is greatly expanded, and the basic attributes of users of the entire Internet can also be analyzed.