Overlapping Statistical Segmentation for Effective Indexing of Japanese Text.

Ogawa, Yasushi; Matsuda, Toru

Notes FAQ Contact Us

Back to results

Peer reviewed

ERIC Number: EJ595380

Record Type: Journal

Publication Date: 1999

Pages: N/A

Abstractor: N/A

ISBN: N/A

ISSN: ISSN-0306-4573

EISSN: N/A

Available Date: N/A

Overlapping Statistical Segmentation for Effective Indexing of Japanese Text.

Ogawa, Yasushi; Matsuda, Toru

Information Processing & Management, v35 n4 p463-80 Jul 1999

Discusses statistical word indexing for Japanese information-retrieval systems and proposes a new method that uses statistics about characters to evaluate a bi-gram's likelihood of being a word boundary. Describes a new segmentation strategy that extracts some overlapping segments and results in higher retrieval effectiveness. (Author/LRW)

Descriptors: Indexing, Information Retrieval, Japanese, Statistical Analysis, Structural Analysis (Linguistics)

Publication Type: Journal Articles; Reports - Descriptive

Education Level: N/A

Audience: N/A

Language: English

Sponsor: N/A

Authoring Institution: N/A

Grant or Contract Numbers: N/A

Author Affiliations: N/A