Created in 2006, LT³ is a dynamic research group at Ghent University's Department of Translation, Interpreting and Communication. LT³ conducts fundamental and applied research in the domain of language and translation technology and has extensive expertise in the use of machine learning for a wide range of language technology problems, int.al. part-of-speech tagging and lemmatization, anaphora resolution, word sense disambiguation and named entity recognition. LT³ is currently very active in the following areas:
- Terminology: Terminology management is crucial both for accurate translation and for intelligent search applications. Whereas CvT-GenTerm, the department's Terminology Centre, focuses on the manual construction of term banks, LT³ has developed a tool that automatically extracts term lists from monolingual and bilingual corpora. Ongoing research projects focus on the automatic detection of synonyms and hyperonyms and terminology extraction from comparable corpora.
- Translation technology: Translation technology has become an integral part of the translation process, and is one of the more recent additions to the LT³ research agenda. Current topics of interest are the comparison of different methods of translation (human vs. post-editing, human vs. CAT), translation quality assessment, and confidence estimation for machine translation.
- Sentiment analysis and subjectivity detection: The actual meaning of a text goes far beyond the word level. LT³ is currently developing methodologies for sentiment analysis and subjectivity detection through the deep semantic analysis of text and deeply annotated corpora. Envisioned applications are the automatic detection of cyberbullying, suicide detection, opinion mining, personalized advertising, and the detection of subjectivity in annual reports.