“My expectations for KIWI are actually very high”, says Pavel Smrz. “From the very begining, even from the project preparation phase, it has been obvious that KiwiEU-funded project combining the wiki philosophy with methods of the Semantic Web, aiming to develop a new approach to knowledge management. (http://www.kiwi-project.eu/index.php/about) could be an exceptional project.” Pavel is an associate professor at the Dept. of Computer Graphics and Multimedia at Brno University of Technology, which is the Czech Republic’s second largest technical university. He already has a number of research projects on his list – with privacy matters being on my permanent radar, I was particularly intrigued by the Caretaker Project which was conducted in collaboration with eight other partners, including Rome’s Mobility Agency and Kingston University, London. Caretaker stands for ‘Content Analysis and REtrieval Technologies to Apply Knowledge Extraction to massive Recording’ – meaning that it aims at the automatic analysis of, for instance, CCTV and surveillance camera recordings. This adds a whole new dimension to video monitoring! Another project Pavel contributed to is AMI: Augmented Multi-party Interaction.
Researchers on Pavel’s team are Marek Schmidt and Petr Knoth, both currently finishing their master’s thesis in Brno’s Intelligent Systems Master program and later proceeding into the doctoral program. Together with Pavel, Marek offered a tutorial on Information Extraction and Ontology Learning from Texts last summer. Petr is writing his thesis on the extraction of semantic relations from text – all three of them are members of the Natural Language Processing Research Group.
Within the KIWI project, Pavel and his team are going to be responsible for Information ExtractionInformation extraction (IE) is a type of information retrieval whose goal is to automatically extract structured information from unstructured machine-readable documents, generally human language texts by means of natural language processing (NLP). Due to the difficulty of the problem, current ..., as part of the Enabling Technologies Section. Semantic wikis do currently not provide support for annotating content – as annotationAn annotation is notes that you make to yourself while you are reading information in a book, document, online record, video, software code or other information, "in the margin", or perhaps just underlined or highlighted passages. Annotated bibliographies, give descriptions about how each source ... can be quite a daunting task, the goal of their contribution is to develop an information extraction component that semi-automatically extracts meta-data from wiki pages and that interactively guides the user through the annotation task. “I believe our team will not only apply advanced information extraction tools in the context of the semantic wiki (which is our primary task), but we will also learn a lot from other academic as well as from industrial partners,” comments Pavel. “It is especialy important for young researchers and I am very glad we could find and bring excellent students to the team.”