Patent · US Active

Using source-channel models for word segmentation

US7493251B2 · kind B2 · utility

136Cited by

5References

51Claims

0Family size

Assignee

Microsoft Corporation · US

Inventors

Jianfeng Gao · Woodinville, US
Mu Li · Beijing, CN
Chang-Ning Huang · Beijing, CN
Jian Sun · Beijing, CN
Lei Zhang · Beijing, CN
Ming Zhou · Beijing, CN

Key dates

Filing date	May 30, 2003
Grant date	Feb 17, 2009
Priority date	—
Expiry date	Jan 9, 2027

Classification

Technology area (CPC G)Physics
CPC primaryG06F40/284
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A method and apparatus for segmenting text is provided that identifies a sequence of entity types from a sequence of characters and thereby identifies a segmentation for the sequence of characters. Under the invention, the sequence of entity types is identified using probabilistic models that describe the likelihood of a sequence of entities and the likelihood of sequences of characters given particular entities. Under one aspect of the invention, organization name entities are identified from a first sequence of identified entities to form a final sequence of identified entities.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.