The French Research groups TIA and
ASSTICCOT, the IRAIA (European project)
consortium and other researchers from corpus linguistics, terminology, AI, knowledge engineering, NLP, information sciences or ML have the willingness to propose a expression of
interest to the European Commission.
The topic would be related to the engineering of semantic knowledge for Information Retrieval or to assist content-based access to documents.
We are setting up this proposal and inviting european organisations that could be interested in it.
A more precise proposal is under writing by Claire Nedellec (from INRA, who works in ML), Adeline Nazarenko from LIPN (NLP), Didier Bourigault, linguist at ERSS and Kurt Englmeier (DIW).
The dead-line for sending such expressions on interest is June 7th 2002, at http://www.cordis.lu/fp6/eoi-instruments/
A short description of the expression of interest follows, so that you could tell us if you would like to join the network of laboratories and companies that will hold this proposal.
If you are interested in this expression of interest, please send the following information to Josiane Mothe, from IRIT :
Building semantic knowledge for an intelligent access to text content
Beside the generalisation of multimedia communication, the volume of textual information is exponentially increasing.
Today mere Information Retrieval technologies are unable to meet specific information needs.
Developing intelligent tools and methods able to give access to text contents is therefore more than never a key issue for knowledge and information management.
From the document engineering system of a small firm to the management of a whole scientific documentation and from an Intranet information system to the exploitation of the semantic web, text content access is a crucial issue.
As soon as one wants to automates the access to content of texts, one need semantic knowledge to localize and interpret the relevant information.
The problem of acquisition of semantic knowledge is a well-known bottleneck for real-world applications, whichever technology is used (Information Extraction, Information retrieval, Question/Answering, and more generally document engineering).
There are two main reasons.
First, much effort has been devoted to generic knowledge, which is difficult to acquire. Until now, for example endeavour has mainly been devoted to the building of general knowledge bases, either lexical database such as WordNet or EuroWordNet or general ontologies (CYC, for instance) and to the definition of formal languages for the representation of ontology.
In contrast, almost no federal effort has been devoted to the specific semantic knowledge that is required for any application.
We argue that no generic knowledge can be used as such and that the required semantic knowledge must be specifically tuned for the application, domain and task, it will be used for.
Although the process of acquiring this specific semantic knowledge cannot be fully automatic, tools can be developed to efficiently help its acquisition.
Second, it is also noticeable that there has been little dialogue between the various disciplines involved in knowledge acquisition and text analysis, although the integration of methods and tools from various disciplines are needed.
Various experiments in this framework have been conducted which yield encouraging results. For instance, in France, a whole community of researchers have emerged around the TIA working group, which has been working for almost 10 years on the acquisition of semantic knowledge from textual data.
Different teams, coming from various research fields (information science, Artificial Intelligence, Linguistics, Natural Language Processing, Machine Learning, Terminology, Knowledge Acquisition…) have been involved in that acquisition process but effort remains scattered.
To go further towards the development of a global and unifying methodology and set of tools for the building of semantic knowledge and its exploitation to give access to text content, it is now crucial to combine these approaches.
