Wikipedia and WordNet integration based on words co-occurrences - Publication - MOST Wiedzy

Search

Wikipedia and WordNet integration based on words co-occurrences

Abstract

The article presents a method for automatic integration of two lexical resources: semantic dictionary WordNet and electronic encyclopaedia Wikipedia. Our goal is to add automatically an semantic tags - a WordNet synset identifier to the title of the Wikipedia article. We've analyze several different ap-proaches to these problem and implement our own solution, based on word occurrences in synsets descriptions and the article body. Application of our algorithm as a result gives Wikipedia articles automatically annotated with WordNet synsets, what gives semantic readability of the knowledge stored in encyclopaedia. The procedure results has been evaluated trough comparison with hand crafted golden standard. At the end of the article we introduce some possible modifications to improve our procedure and reach higher precision of disambiguation Wikipedia articles.

Full text

Details

Category:
Monographic publication
Type:
rozdział, artykuł w książce - dziele zbiorowym /podręczniku w języku o zasięgu międzynarodowym
Title of issue:
Information systems architecture and technology : advances in web-age information systems strony 93 - 103
Language:
English
Publication year:
2009
Bibliographic description:
Kilanowski J., Szymański J.: Wikipedia and WordNet integration based on words co-occurrences// Information systems architecture and technology : advances in web-age information systems/ ed. (eds.) L. Borzemski, A. Grzech, J. Świątek, Z. Wilimowska. Wrocław: Oficyna Wydawnicza Politechniki Wrocławskiej, Wrocław, 2009, s.93-103
Verification:
Gdańsk University of Technology

seen 16 times

Publikacje, które mogą cię zainteresować

Meta Tags