By Gregory Grefenstette (auth.), Gregory Grefenstette (eds.)
Most of the papers during this quantity have been first offered on the Workshop on Cross-Linguistic details Retrieval that used to be held August 22, 1996 dur ing the SIGIR'96 convention. Alan Smeaton of Dublin collage and Paraic Sheridan of the ETH, Zurich, have been the 2 different individuals of the medical Committee for this workshop. SIGIR is the organization for Computing Ma chinery (ACM) distinct curiosity workforce on details Retrieval, and so they have held meetings every year for the reason that 1977. 3 extra papers were extra: bankruptcy four allotted Cross-Lingual details retrieval describes the EMIR retrieval method, one of many first common cross-language structures to be carried out and evaluated; bankruptcy 6 Mapping Vocabularies utilizing Latent Semantic Indexing, which initially seemed as a technical file within the Lab oratory for Computational Linguistics at Carnegie Mellon college in 1991, is integrated the following since it used to be one of many earliest, notwithstanding hard-to-find, publi cations displaying the appliance of Latent Semantic Indexing to the matter of cross-language retrieval; and bankruptcy 10 A Weighted Boolean version for pass Language textual content Retrieval describes a contemporary method of fixing the interpretation time period weighting challenge, particular to Cross-Language details Retrieval. Gregory Grefenstette members Lisa Ballesteros David Hull W, Bruce Croft Gregory Grefenstette heart for clever Xerox examine Centre Europe details Retrieval Grenoble Laboratory computing device technology division collage of Massachusetts Thomas okay. Landauer division of Psychology Mark W. Davis and Institute of Cognitive technology Computing study Lab college of Colorado, Boulder New Mexico nation college Michael L. Littman Bonnie J.
Read Online or Download Cross-Language Information Retrieval PDF
Best nonfiction_7 books
- Fuel Cell Technology: Reaching Towards Commercialization
- Successful IT Outsourcing: From Choosing a Provider to Managing the Project
- Physique Des Solides [physics]
- Bioinformatics - Trends and Methodologies
Extra info for Cross-Language Information Retrieval
3 EXPERIMENTATION ON A LIBRARY CATALOG The catalog of our libraries is done by downloading OCLC notices that are updated to add local information. In this catalog, titles of documents are mainly in French and in English. The same catalog entry can have a list of keywords in English and another in French but they are not the direct translation of each other. This database is really a mixed language database in the sense that in a same entry there can be more than one language. For this reason, we adopted this database to experiment the architecture described in the previous section.
5 AUTOMATIC CROSS-LANGUAGE INFORMATION RETRIEVAL USING LATENT SEMANTIC INDEXING Michael L. Littman Susan T. Dumais* Thomas K. Landauer** Computer Science Department Brown University * Information Sciences Resew'ch Group Bellcore, Morristown, NJ ** Department of Psychology and Institute of Cognitive Science University of Colorado, Boulder 1 INTRODUCTION We describe a method for fully automated cross-language document retrieval in which no query translation is required. Queries in one language can retrieve documents in other languages (as well as the original language).
For shorter queries, the method also reduces the error due to failure to accurately translate phrases and specialized vocabulary. For both the ES-BASE and SE-BASE queries, those queries that showed improvement via pre-translation alone gained greater improvements from subsequent post-translation feedback. This suggests that the pre-translation feedback stage creates a better base for translation and then the post-translation stage reduces the negative effects of ambiguity caused by inappropriate term definitions.
Cross-Language Information Retrieval by Gregory Grefenstette (auth.), Gregory Grefenstette (eds.)