For a list of University of Antwerp projects, see my official UA research projects page

Current Research Projects

Genre classification for Document Management (DAC, South Africa) (cooperation with Trifonius, CSIR and CTexT).

ALADIN: Adaptation and Learning for Assistive Domestic Vocal Interfaces (IWT-SBO).' Using grammar induction to learn to adapt voice interfaces to persons with physical impairments.

Biograph (BOF-GOA). Text Mining with Heterogeneous Databases. Combining text mining and datamining with heterogeneous information sources in biomedical discovery and in knowledge management.

Starling (FWO). Application of Statistical Relational Learning of Natural Language to problems in natural language processing (event extraction, semantic role labeling, coreference resolution).

Daphne (IOF). “A safer internet: semi-automatically recognizing internet pedophilia in multilingual online social networks.” Development of tools for the stylometric analysis of language use by pedophiles.

STYLENE (EWI). A web service for stylometry and readability research for the Dutch language.

TTNWW (EWI). TST tools for Dutch as web services in a workflow.

Delearyous (IWT). Interpersonal communication training of natural language interaction with autonomous virtual characters. Development of emotion and sentiment text analysis.

Audio Description in Dutch: corpus research for a new multimodal translation form (BOF, Artesis). (Cooperation with Aline Remael).

Analysis of the impact of news on Belgian share prices through text mining (National Bank of Belgium). (Cooperation with David Martens).

Authorship, composition and textual interconnectedness of three 16th-century mystical texts. Die evangelische peerle, Vanden tempel onser sielen, and the Arnhem mystical sermons. A stylometric approach' (FWO). Stylometric research on 16th century texts. (Cooperation with Kees Schepers).

Abstract rules or statistical learning? The impact of lexical and sublexical homophony in spelling and reading homophonous verb forms (FWO). (Cooperation with Dominiek Sandra).

Some Past Research Projects

AMICA: Automatic monitoring for cyberspace applications (IWT-SBO). Preparing research on semi-automatic monitoring of social networks.

Computational Stylometry for Dutch (FWO). Combining machine learning and automatic text analysis for author recognition and style analysis.

Development of Dutch resources for UBISOFT's My Word Coach.

DAESO (Stevin, Nederlandse Taalunie). Development of a summarization engine using semantic overlap detection tools.

GRAVITAL (IWT). Developing a natural language interface to a graphical package combining shallow parsing and expert system rules.

COREA (Stevin, Nederlandse Taalunie). Development of a coreference resolution system for Dutch based on Machine Learning and its evaluation in the context of information extraction and question answering.

IMIX-ROLAQUAD (NWO). Basic research in the context of a multiple-turn question-answering dialogue application for Dutch.

BioMinT (EU). Development of text analysis techniques for medical language.

FLAVOR (IWT). Morpho-phonological and morpho-syntactic models for speech recognition for Dutch.

Ontobasis (IWT). Using shallow parsing for the extraction of ontologies from text.

MUSA (EU). Using text analysis for automatic condensation of speech-recognized text to subtitles and as input to translation (English speech input).

ATRANOS (IWT). Using text analysis for automatic condensation of speech recognition transcripts into subtitles (Dutch input).

PROSIT (FWO and NWO). Using machine learning and automatic text analysis for prosody generation in text to speech.

PASCAL (EU). Research network on machine learning (text analysis and computational modeling of acquisition themes).

CALCUA (Flemish funding). Using a supercomputer for solving optimization problems in machine learning of language.