This paper presents an algorithm for unsupervised noun sense induction, based on clustering of Web search results. The algorithm does not utilize labeled training instances or any...
Goldee Udani, Shachi Dave, Anthony Davis, Tim Sibl...
The purpose of this paper is to automatically create multilingual translation lexicons with regional variations. We propose a transitive translation approach to determine translat...
: When one tries to use the Web as a dictionary or encyclopedia, entering some single term into a search engine, the highly-ranked pages in the result can include irrelevant or use...
This paper describes a visualisation tool, VlUM, designed to support users in scrutinising models of their interests, preferences and knowledge. We also describe MECUREO, a tool f...
We present two methods for estimating replacement probabilities without using parallel corpora. The first method proposed exploits the possible translation probabilities latent in ...