Computational lexicon for Maltese : lessiku komputazzjonali Malti

Abstract

Lexicon Structuring Technique (LST) has been applied to a large corpus of Maltese text to create the first large scale Computational Lexicon for Maltese. The system was thought to disambiguate between ambiguous Maltese words through human interaction. A web-based client-server implementation is presented together with an extensive discussion of the results obtained. The client-server implementation handles interoperability with other Language Resources through innovative use of technology such as XML, SOAP and UDDI and the use of international markup standards such as XCES-EAGLES. The results obtained as part of this work from the foundation of a future wide range of computational linguistics tools for Maltese that will enable the whole field of Maltese Computational Linguistics to advance significantly while facilitating the setup of the Maltese National Corpus. Future application of LST to other languages also looks promising due to technique’s superiority in handling languages of both Indo-European and Semitic origin.