ILPnet2: ILP Network of Excellence
(NoE Nr. 977102)

Progress Report for the Period 01/09/1999 to 31/08/2000

Background, objectives and workplan for the second year

ILPnet2 is the Inductive Logic Programming (ILP) Network of Excellence consisting of over 20 universities and research institutes. In addition the network actively pursues industrial relations through its End-User Club (EUC), consisting of companies and other non-academic institutions interested in practical applications of ILP. ILPnet2 succeeds ILPnet, the ILP European Scientific Network. The mission of ILPnet2 is defined by the following long-term objectives: The workplan for the second year was as follows:

Activities in the second year

Below we report on the activities of ILPnet2 in the second year and relate them to the workplan. These include the publication of the ILP Newsletter, significant updates of the WWW resources (library, systems and application descriptions) and the organization of a number of educational events, one of which was addressed at the EUC. More detailed information on ILPnet2 activities can be found starting at the ILPnet2 website at http://www.cs.bris.ac.uk/~ILPnet2/.

Educational events and workshop

In the second year, ILPnet2 supported a number of ILP-related educational events. These were as follows: The 10th International Conference on Inductive Logic Programming (ILP-2000), also supported by ILPnet2, took place 24-27 July 2000 in London. It was integrated with the First International Conference on Computational Logic (CL-2000). This exposed the ILP community to a wide range of work in computational logic and logic programming and at the same tive gave ILP researchers the opportunity to present their achievements to the computational logic community. A report on ILP-2000 is included in the appendix.

Common electronic workspace

The common electronic workspace for computer supported cooperative work was created in year one. In the second year, we collected educational materials from the ILPnet2 supported educational events listed above and put them in the electronic workspace. The materials in the electronic workspace are available to ILPnet2 nodes and EUC members. Some additions were also made to the publicly available list of educational materials. on the web. The educational materials in the electronic workspace at the end of year two are the following:

ILPnet2 website

ILPnet2 exploits the Internet as the main medium of information dissemination. Information disseminated in this way includes bibliographic information (also technical reports and thesis abstracts), information on ILP systems and information on ILP applications and datasets. In this respect, ILPnet2 expands on activities undertaken by ILPnet, the ILPnet2 predecessor. In year two, the ILPnet2 on-line library expanded from 750 entries to almost 900 entries. In addition, a further 300 entries have been collected from various sources, such as on-line websearches, library searches, etc. Once these entries have been added to the database, the library will have doubled in size compared to the start of ILPnet2. The ILPnet2 systems page, which was set up during the first year, has been updated in several ways. First, pointers to WWW pages of the ILP systems have been added. Thus, each entry in the list of ILP systems contains at least one of two possible pointers. One is a pointer to a short description of the system (the format of all short descriptions is uniform) and another one points to the WWW page of the system, usually provided by the developers. Second, eight new entries have been added to the list, seven pointers to WWW pages and one short description. The ILPnet2 applications and datasets page has been restructured. The pointers to short descriptions of applications (with common format) have been grouped in five sections according to their application domains: engineering, life sciences, environmental sciences, natural language processing and other. Also, twenty-five new short descriptions have been added in the last year. The ILPnet2 website also contain information on the portability of ILP systems to different platforms. This activity began ahead of schedule in year one (four systems tested) and continued in year two (three systems tested). Other information available through the ILPnet2 website includes the ILP newsletter, a list of ILP-related events and a list of educational materials. The ILP Newsletter had four issues in the second year of ILPnet2. In addition to being distributed via email, the newsletter is also available through the ILPnet2 website at IJS. A list of ILP-related events was also created in year two, containing 20 events for that year. The list of educational materials, which include on-line versions of lectures/slides and syllabi of ILP or ILP related courses, was updated with several new entries.

End-user events

Two events addressed at end-users took place in year two: a seminar on data analysis in life sciences and a data analysis workshop. A two-day seminar on Data Analysis in Life Sciences was held in Ljubljana, Slovenia, 18-19 May 2000. Lecturers included Ashwin Srinivasan, Ross King and Saso Dzeroski. A substantial part of the seminar was devoted to ILP and its applications in life sciences (e.g., to drug design and analysis of genome data). This seminar was attended by six persons from two institutions in the ILPnet2 EUC: The National Institute of Chemistry (Ljubljana) and DIALOGIS (Bonn). Next to the seminar, a data analysis workshop was organized. Ashwin Srinivasan and Saso Dzeroski worked together with Kathy Astrahantseff (DIALOGIS) to analyze data from protein interaction assays. Three data analysis tasks were addressed, including the detection of motifs common to a set of proteins, a typical ILP task. The initial results were encouraging and the domain expert is now collecting more data from similar protein interaction assays. A more detailed report is given in the appendix.

Expansion of the network

This was a key concern in the second year of the network. 3 new CEE nodes have been added to the network: the Technical University of Kosice (SK), the Rudjer Boskovic Institute in Zagreb (HR), and the Masaryk University in Brno (CZ). In addition, 5 new associate nodes (this is a new category created for new West-European nodes) have been added: Albert-Ludwigs University Freiburg (D), Otto-von-Guericke University Magdeburg (D), Oxford University (UK), Universidade Nova de Lisboa (P), and University of Wales, Aberystwyth (UK). Finally, one new end-user has been added to the end-user club: Swiss-Life Insurance in Zürich (CH). Overall, the network now consists of 25 nodes, 5 associate nodes, and 14 end-users.

Summary and outlook

In summary, all of the activities planned in the workplan for the second year have taken place in their originally planned or a similar form. The planned problem presentation workshop for end users was replaced with a data analysis workshop (planned for year three). This makes sense, as problem presentations already took place at the seminar in Caminha (year one). We still plan to have at least one data analysis workshop in year three. We did not have a separate summer school on ILP and its applications, but had a variety of other educational events. In particular, the two ILP-related courses at ESSLI-2000 can be considered as a replacement for an ILP summer school: the two courses comprised 20 hours of lectures. In general, we consider the strategy of giving ILP-related courses at larger established events as a better method of attracting a wider audience. In addition to continuing dissemination of various kinds of information through the WWW, activities planned for the second year include the organization of an ILP workshop, a summerschool on ILP and its applications, and the preparation/dissemination of an ILP toolbox for educational purposes.

Appendix: Reports on Events Sponsored by ILPnet2

About this document ...

This document was generated using the LaTeX2HTML translation Version 99.1 release (March 30, 1999)