Computational Linguistics Group Department of Computer Science University of Haifa Home People Projects Publications Events

Project description

Objective

To create a Hebrew WordNet, aligned with the Princeton English WordNet.

Researchers

In Haifa, Noam Ordan, Iris Eyal and Shuly Wintner (and, in the past, Margalit Zabludowski). This project is joint with a team at the TCC Division ITC-IRST, Trento, Italy.

Status

Complete

Funding

Israeli Ministry of Science and Technology, as part of the Knowledge Center for Hebrew Language Telecommunication

Abstract

Following the success of the English WordNet, similar networks have been developed for a variety of languages. In particular, researchers at ITC-IRST in Trento, Italy have developed a methodology for parallel development of multilingual WordNets. The system, called MultiWordNet, contains information on several aspects of multilingual dictionaries, including lexical relationships between words, semantic relations over lexical concepts, several mappings of lexical concepts in different languages etc. See this paper for more information on the methodology.

MultiWordNet has a variety of applications, including:

• Information retrieval: lexical relations can significantly improve the performance of query answering systems, for example; multilingual relationships facilitate multilingual information extraction and retrieval.
• Semantic annotation: since words in the network are tagged by the semantic concepts to which to relate, a multilingual WordNet can be used for semantic annotation and classification of texts.
• Disambiguation: semantic relationships can assist in determining the semantic distance between words and concepts, thereby assisting in lexical disambiguation.
• Terminology: the system can be used for developing structured terminologies for specific applications.

Our goal in this project is to use the MultiWordNet methodology in order to construct a Hebrew WordNet, aligned with the one developed at IRST (and, therefore, aligned with English, Italian and Spanish).

Resources

• A graphical user interface for accessing the Hebrew network is available under MultiWordNet on-line.
• Entry point for developers (password required).
• Download the Hebrew netowrk. Access to and use of these products is governed by a license. If you use this resource for academic work, please cite the following paper.

Publications

• Noam Ordan and Shuly Wintner. Hebrew WordNet: a test case of aligning lexical databases across languages. International Journal of Translation 19(1):39-58, 2007. PDF
• Noam Ordan and Shuly Wintner. Representing natural gender in multilingual lexical databases. International Journal of Lexicography 18(3):357-370, September 2005. PDF.

Contact