Authors:
Theodoros Alevizos
1
;
Vassilis G. Kaburlasos
1
;
Stelios Papadakis
1
;
Christos Skourlas
2
and
Petros Belsis
2
Affiliations:
1
T.E.I. of Kavala, Greece
;
2
T.E.I. of Athens, Greece
Keyword(s):
Fuzzy Interval Number (FIN), Information Retrieval (IR), Cross Language Information Retrieval (CLIR).
Related
Ontology
Subjects/Areas/Topics:
Advanced Applications of Fuzzy Logic
;
Artificial Intelligence and Decision Support Systems
;
Enterprise Information Systems
;
Industrial Applications of Artificial Intelligence
Abstract:
Fuzzy Interval Numbers (FINs) could be seen as a set of techniques applied in Fuzzy System applications. In this paper, we propose a series of techniques to solve multi-Lingual and Cross Language Information Retrieval (CLIR) problems, based on Fuzzy Interval Numbers (FINs). Some experiments showing the importance of these techniques in the CLIR-systems are briefly described and discussed. Our method is evaluated using monolingual and bilingual public bibliographic data extracted from the National Archive of the Greek National Documentation Centre. All the experiments were conducted with and without the use of stemming, stop-words and other language dependent (pre-) processing techniques. It seems that a main advantage of our approach is that the method is language independent and there is also no need for any text pre-processing or higher level processing, avoiding thus the use of taggers, parsers, feature selection strategies, or the use of other language dependent NLP tools.