06904402 is referenced by 77 patents and cites 3 patents.

A method for optimizing a language model is presented comprising developing an initial language model from a lexicon and segmentation derived from a received corpus using a maximum match technique, and iteratively refining the initial language model by dynamically updating the lexicon and re-segmenting the corpus according to statistical principles until a threshold of predictive capability is achieved.

Title
System and iterative method for lexicon, segmentation and language model joint optimization
Application Number
9/609202
Publication Number
6904402 (B1)
Application Date
June 30, 2000
Publication Date
June 7, 2005
Inventor
Lee Feng Chien
Beijing
CN
Dong Feng Cai
Liaoning Province
CN
Jianfeng Gao
Beijing
CN
Shuo Di
San Mateo
CA, US
Kai Fu Lee
Woodinville
WA, US
Chang Ning Huang
Beijing
CN
Hai Feng Wang
Hong Kong
CN
Agent
Microsoft
Assignee
Microsoft Corporation
WA, US
IPC
G10L 015/00
G06F 017/20
G06F 017/21
View Original Source