ILK: Induction of Linguistic Research Group - Home Page
ILK: Induction of Linguistic Knowledge Research Group

The ILK research group is part of the Tilburg centre for Cognition and Communication (the Creative Computing research programme) and the Department of Communication and Information Sciences of the Faculty of Humanities (Dutch name: Faculteit Geesteswetenschappen) of Tilburg University, The Netherlands.

ILK shares much of its history with, and remains closely related to the Language Technology Research group of CLiPS (formerly CNTS, the Center for Dutch Language and Speech) of the University of Antwerp, Belgium.

Originating partly from ILK research are Textkernel, an Amsterdam-based company developing tools for information extraction, text understanding, and content management, and STIL, the Foundation for Inductive Learning Applications, our consultancy and software license broker.

Research in ILK has mostly been funded by public research funds. Our main sponsors are NWO, The Netherlands Organisation for Scientific Research, and SenterNovem (the Dutch Ministry of Economic Affairs).


Legacy ILK web pages

A brief history

ILK, Induction of Linguistic Knowledge, came into existence in the early nineties within the Institute for Language and Artificial Intelligence (ITK) at Tilburg University, when Walter Daelemans enthused a group of students to work on the new concept of applying recently developed machine learning algorithms to problems in natural language processing. ILK got its name from the six-person 1995-2001 NWO/TSL project awarded to Walter Daelemans. Since then it has continued to be the flag name of a string of projects in which inductive learning algorithms are developed and employed in solving natural language problems. The research was, and continues to be, intended to be useful in two areas:

  • Language Engineering. Provided there is material to learn from, inductive learning techniques can bypass the knowledge acquisition bottleneck that plagued computational linguistics before; they are now widely employed across a wide range of NLP tasks, producing generally accurate, robust, practical, language-independent methods and systems.

  • Linguistic and Cognitive Modelling. As a new method in linguistic analysis, inductive learning allows us to discover linguistic knowledge in corpora and dictionaries. They also provide a new perspective on old philosophical and cognitive science problems concerning the representation and acquisition of linguistic knowledge.

The book "Memory-Based Language Processing" (Daelemans and Van den Bosch, 2005) documents the first fifteen years of work of the Tilburg and Antwerp groups on applying memory-based classification to natural language processing tasks. | Last update: