Approaching Analysis of EU IST projects database

We present the first results of the analysis of publicly available EU IST project descriptions. The database was automatically built from the publicly available information on the Web and organized to enable further analysis. We have used Text Mining methods to group the projects according to their content and the institutions participating in the projects. Two systems were developed, the first enabling grouping and visualization of the projects and the second enabling searching over the projects and partners. One of the advantages of the proposed methodology is that it is scalable and can be easily applied on large collections of project descriptions. By changing the filters needed for automatic creation of the database from the Web site, the approach could be used on other Web sites giving similar type of information as EC site with funded project descriptions. Our intention here is mainly to show potential of Text Mining and stimulate further development of application oriented Text Mining systems. Our customers here are EC officers and analysts in charge of EU funding of projects. The first feedback we got from them on two presentations we had was very positive, contributing many ideas for potential problems and improvements of the first version of the system.