US 7,617,090 B2
Contents filter based on the comparison between similarity of content character and correlation of subject matter
Jiang Wang, Beijing (China); Jianzhong Gao, Beijing (China); Nan Wang, Beijing (China); Guang Zhu, Beijing (China); and Hang Xiao, Beijing (China)
Assigned to Legend (Beijing) Limited, Beijing (China)
Appl. No. 10/488,731
PCT Filed May 23, 2002, PCT No. PCT/CN02/00346
§ 371(c)(1), (2), (4) Date Mar. 05, 2004,
PCT Pub. No. WO03/038667, PCT Pub. Date May 08, 2003.
Claims priority of application No. 01 1 31420 (CN), filed on Sep. 07, 2001.
Prior Publication US 2004/0243537 A1, Dec. 02, 2004
Int. Cl. G06F 17/27 (2006.01); G10L 11/00 (2006.01)
U.S. Cl. 704—9  [704/270; 704/270.1] 41 Claims
OG exemplary drawing
 
1. A contents filter based on similarity of content character and correlation of subject matter, which is characterized in that the contents filter includes at least a filtering system and a disciplining system wherein said filtering system and the disciplining system are installed physically separately, and the filtering system is installed in at least one input device of network information and communicates with the disciplining system through a data interface;
the disciplining system learns with appointed information to obtain filtering characters of said appointed information;
the filtering system filters said appointed information, and the disciplining system communicates with the filtering system;
said disciplining system includes an anti-interference extracting module of text character for contents filtering, the module finds a specified text information in a checked text to determine whether a sequence of the specified text contents is in accord with a sequence of a preset text wherein different filtering characters that are obtained by the disciplining system are configured to filtering systems located in different input devices of network information; and thereby determines an interferential distance between the specified text information and the checked text, if the interferential distance is less than a preset threshold, the checked text contents are set as the interferential text contents to be selected, wherein said configuration is to distribute the filtering character of the filtering system according to burden capacity, location and purpose of the input device of network information in network.