!!!This is an old page, kept for reference only!!!
I am currently working as research fellow at the University of Sheffield (since beginning of 2013)
I am currently working as research fellow at the University of Sheffield (since beginning of 2013)http://www.dcs.shef.ac.uk/cgi-bin/makeperson?J.Petrak
Austrian Reseach Institute for Artificial Intelligence
Freyung 6/6, A-1010 Vienna, Austria
Tel: +43-1-533 6112 17
Fax: +43-1-533 6112 77
Email: johann[DOT]petrak(AT)ofai[DOT]at (GPG public key)
Semantic systems and Ontologies
Language Technolog and Natural Language Processing
Machine Learning and Data Mining
Current interests: Ontology-based information extraction from natural language texts, domain-specific knowledge extraction from websites, semantic annotation and search of web and desktop resources
Current and recent projects
OREX: Ontology-Based Information Extraction and Search
SEMPRE: Semantically Aware Profiling for Recommenders
Aurex/W: Automated Retrieval and Extraction of Web Contents
BioMint: "Biological Text Mining"
METAL: "A Meta-Learning Assistant for Providing User Support in Data Mining and Machine Learning"
Sol-EU-Net: "Data Mining and Decision Support for Business Competitiveness: A European Virtual Enterprise" (EU IST Project)
Plugins and software for GATE:
= AppDoc a GATE plugin for documenting GATE applications
= StringAnnotation a GATE plugin that provides the ExtendedListGazetteer and SimpleRegexpAnnotator PRs
= JapeUtils a GATE plugin that provides the JapeExtended PR which, among other things, makes it possible to check for feature value equality between annotations
= Perl GATE a perl module for handling GATE documents
= VirtualDocuments a GATE plugin that allows the on the fly creation and processing of documents from their annotation content = VirtualCorpus a GATE plugin that provides a corpus backed by files in a directory or fields in a JDBC table
MLEE: a machine learning experimentation environment developed for the METAL project.
C45OFAI: A patch that extends C4.5 with several new features.
BASECLEARN: A fast baseline classification learning algorithm
Dynamically created list of current