Vol. 9, No.1/2, May 2008
 
Vlado Kešelj
Dalhousie University
 
Danko Šipka
Arizona State University
 

A SUFFIX SUBSUMPTION-BASED APPROACH TO BUILDING STEMMERS AND LEMMATIZERS FOR HIGHLY INFLECTIONAL LANGUAGES WITH SPARSE RESOURCES

 
Abstract: We present a general suffix-based method for construction of stemmers and lemmatizers for highly inflectional languages with only sparse resources. The process is directly implementable with described efficient design and it is evaluated on a construction of a stemmer for the Serbian language. The evaluation on real data has shown an accuracy of 79%.

PDF

 


ARTICLE

Biljana Kosanović 
ACCESSING SCIENTIFIC INFORMATION IN SERBIA: SIX YEARS EXPERIENCE

Cvetana KrstevBojana DjordjevićSanja AntonićNevena Ivković-BerčekZorica ZoricaVesna CrnogoracLjiljana Macura
COOPERATIVE WORK IN FURTHER DEVELOPMENT OF SERBIAN WORDNET

Miroslav Martinović 
TRANSFER OF NATURAL LANGUAGE PROCESSING TECHNOLOGY: EXPERIMENTS, POSSIBILITIES AND LIMITATIONS CASE STUDY: ENGLISH TO SERBIAN

Vlado KešeljDanko Šipka
A SUFFIX SUBSUMPTION-BASED APPROACH TO BUILDING STEMMERS AND LEMMATIZERS FOR HIGHLY INFLECTIONAL LANGUAGES WITH SPARSE RESOURCES

Duško VitasGordana Pavlović-Lažetić
RESOURCES AND METHODS FOR NAMED ENTITY RECOGNITION IN SERBIA

Ivan ObradovićRanka Stanković
SOFTWARE TOOLS FOR SERBIAN LEXICAL RESOURCES

REVIEWS

Aleksandra Vraneš 
FROM THE HISTORY OF THE LIBRARY AND INFORMATION SCIENCE DEPARTMENT OF THE FACULTY OF PHILOLOGY OF THE UNIVERSITY OF BELGRADE

Aleksandra Nastić
DR NEDELJKO PAREZANOVIĆ, RETIRED UNIVERSITY PROFESSOR

Stela Filipi-Matutinović
10TH INTERLENDING AND DOCUMENT SUPPLY CONFERENCE: RESOURCE SHARING FOR THE FUTURE - BUILDING BLOCKS TO SUCCESS