DOC PREVIEW
UT INF 385Q - Cross-Language IR: An Overview

This preview shows page 1-2-3 out of 8 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 8 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Cross-Language IR: An OverviewWhat is CLIR?CLIR vs. MLIRWhy do we need CLIR systems?General Issues With CLIRBasic Approaches to CLIRBasic Approaches Cont’d…ConclusionsCross-Language IR: An OverviewBernardine ReitmeyerWIRED12/2/04What is CLIR?IR on a monolingual document collection that can be queried in multiple languagesIR on a multilingual document collection, where queries can retrieve documents in multiple languagesIR on multilingual documents(Hull & Grefenstette, 1996)CLIR vs. MLIRCross-language information retrieval refers to the 3 definitions we’ve discussed.Multilingual information retrieval (MLIR) includes CLIR and IR in systems that use a language other than that of the searcher (query must also be in the other language)Most people still use MLIR to describe any work in this areaEmphasis here is cross-languageWhy do we need CLIR systems?To search collections containing documents in many languagesTo search documents containing text in more than one languageFor those who aren’t fluent in a foreign language but can make use of some contents of documents retrieved (images, names, etc.)(Oard & Dorr, 1996)General Issues With CLIRMultilingual text access (character sets, etc.)Differences between languages-stemming, compound words, breaks between words, etc.Term ambiguity between languagesWhen to translate (query vs. document)Basic Approaches to CLIRMachine Translation/Machine Learning(automatic translation of documents or queries)Controlled Vocabulary(use of multilingual thesaurus)Dictionary-Based Approaches(combination of monolingual or bilingual dictionaries – similar to a thesaurus)Basic Approaches Cont’d…Latent Semantic Indexing(makes comparisons between semantically related words)Corpora-Based Approaches(uses existing texts to automatically build information about relationships between terms)ConclusionsAlthough CLIR faces many tough issues, it is an important and growing facet of IR. Starting in 2000, it even got its own conference (because it was getting lots of attention at the TREC conferences) called CLEF (Cross-Language Evaluation Forum).For more information:


View Full Document

UT INF 385Q - Cross-Language IR: An Overview

Documents in this Course
Agents

Agents

12 pages

Groupware

Groupware

20 pages

Load more
Download Cross-Language IR: An Overview
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Cross-Language IR: An Overview and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Cross-Language IR: An Overview 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?