Final Activity Report Summary - MULTILINGUA (Multilingual knowledge tools)
MULTILINGUA addresses research needs for the multilingual information society. It has provided early stage training in the area of language resources and knowledge tools. This implies an interdisciplinary area of study and research where fellows develop knowledge and skills that couple linguistic proficiency with advanced information technology. The training programme has taken advantage of the University of Bergen's status as a centre of research and repository of digital resources and tools aimed at text technologies and multilingual applications, including computational and corpus linguistics, machine translation, term banks, electronic editions and e-books, electronic archives and libraries, multilingual word nets, document management and mark-up, information retrieval and summarisation, etc. Fellowships were awarded to nine fellows from seven European countries.
The research carried out by MULTILINGUA fellows had the following themes:
- Evaluation of Natural Language Parsers
- The Latin Language in the Inscriptions from Roman Dacia
- Electronic Text Technology in Information Society: From Hypertext to Computational Linguistics
- Exploiting ontologies, multilinguality, coherence and advanced machine learning techniques for applied word sense disambiguation
- Contrastive Analysis and Translation English-Spanish: functions of the English -ing form and its equivalents in Spanish
- Obtaining semantic relations from text corpora in an unsupervised way
- Application of the Semantic Mirrors method to the data of a bilingual English-Greek corpus
- Translation of GMO texts from English to Spanish: A multilingual study of terminology and phraseology by means of encoding, parallel alignment and computer analysis of a bilingual corpus of English popular science books and their Spanish translations
- Selectional Preference Acquisition. A preliminary contrastive study on Italian and English verbs.
These projects were successfully carried out and resulted in research results incorporated in PhD dissertations and disseminated through other channels.
The research carried out by MULTILINGUA fellows had the following themes:
- Evaluation of Natural Language Parsers
- The Latin Language in the Inscriptions from Roman Dacia
- Electronic Text Technology in Information Society: From Hypertext to Computational Linguistics
- Exploiting ontologies, multilinguality, coherence and advanced machine learning techniques for applied word sense disambiguation
- Contrastive Analysis and Translation English-Spanish: functions of the English -ing form and its equivalents in Spanish
- Obtaining semantic relations from text corpora in an unsupervised way
- Application of the Semantic Mirrors method to the data of a bilingual English-Greek corpus
- Translation of GMO texts from English to Spanish: A multilingual study of terminology and phraseology by means of encoding, parallel alignment and computer analysis of a bilingual corpus of English popular science books and their Spanish translations
- Selectional Preference Acquisition. A preliminary contrastive study on Italian and English verbs.
These projects were successfully carried out and resulted in research results incorporated in PhD dissertations and disseminated through other channels.