Abstract | ||
---|---|---|
The paper examines issues related to the development of a cross-language information extraction system using a consistent translingual parsing mechanism. Pattern formulation that relies on shallow parsing and relative positioning of terms, while effective in monolingual systems operating on English texts, is all but impossible for languages with free word order. Effective cross-language information extraction necessitates the use of patterns expressed in terms of deep syntactic structures that can be aligned across language boundaries, such as sentence frames and case roles, rather than verbs, noun phrases, and surface-syntactic dependencies. The Meaning Text theory provides a consistent language-independent framework for cross-lingual pattern formulation and matching. |
Year | Venue | Keywords |
---|---|---|
2005 | MLMTA '05: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON MACHINE LEARNING MODELS TECHNOLOGIES AND APPLICATIONS | information extraction,machine translation |
Field | DocType | Citations |
Computer science,Information extraction,Natural language processing,Artificial intelligence,Parsing,Syntax,Relationship extraction | Conference | 0 |
PageRank | References | Authors |
0.34 | 1 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Konstantin Bogatyrev | 1 | 0 | 0.68 |
Roman Yangarber | 2 | 411 | 62.85 |