Fabio Crestani reports on the Third European Summer School in Information Retrieval (ESSIR'2000) held in Varenna, Italy, by Lake Como, on 11-15 September 2000. The event was jointly organised by Maristella Agosti of the University of Padova (Italy), Gabriella Pasi of ITIM-CNR (Italy), and the author of this report who is working at the University of Strathclyde (Scotland).
Here is a story that Fabio Crestani has been telling since his return to Scotland after co-organising ESSIR'2000:
"Coffee or tea?", the young female waitress from the catering service asked me. "Coffee, please", I said, as any real Italian would, even after a few years spent in a tea-drinking country. She served me a good coffee and a good smile. I smiled back politely and that, I think, broke the ice. She had been providing catering services to Villa Monastero, a conference centre partially owned by the Italian National Research Council (CNR), for some time, but something in her look told me that this event was different. As I was one of the organisers of the event, I decided to enquire about this, fearing that something could be wrong. "Nothing wrong at all", she told me, "Just, we are not used to have so many young people attending, here", she added. She explained me that most of the events (conference, workshops, summer schools, and so on) organised at Villa Monastero were attended by somewhat "older" people. The average age of participants to this event was considerable lower than that of other events organised there, she intended. "What topic is it, that you are teaching, that attract such a young crowd?", she asked me. "Information Retrieval", I said puzzled.
For those of you that are not familiar with this topic, like the young waitress, Information retrieval (IR) is the science and technology concerned with the effective and efficient retrieval of information by its semantic content [1]. The central problem in IR is the quest to find the set of relevant documents, amongst a large collection, containing the information sought, thereby satisfying an user's information need usually expressed by a natural language query. Documents may be objects or items in any medium, text, image, audio, or, indeed a mixture of all three.
IR is certainly not a young topic, having been studied for long time by librarians and computer scientists for over 50 years. However, IR is has had a surge of interest in recent times because of the World Wide Web. Finding information on the Web or in one of the increasingly available digital libraries has been compared to "finding a needle in a haystack" [2] and new technologies and tools need to be designed and developed to make all this information available and really useful to users worldwide.
Access to information has gone through a slow but steady process to adapt to the growth of availability of electronically stored information. When libraries were small, access to a piece of information could be achieved by asking the librarian, a "wise sage" who was supposed to have read every book in the library. The librarian could tell you which book contained the information you needed and where the book was located. When the number of books began to exceed the limits of human memory, categorisation became necessary and library classification systems such as the Dewey or the Library of Congress' were developed. Each book was assigned a set of subject headings that identified the topics treated in the book and a location in the library. Only by knowing the appropriate set of subject headings that identified the searched information one could find the location of the book in the library. With computers and the availability of electronic text comes the possibility of searching through the entire text of documents (book, articles, etc.) to find words and phrases that identify a document as containing the information sought. This free text searching ability meant that the searcher did not have to rely on someone else looking for documents for him or assigning documents to particular categories. Nevertheless, if on one hand this puts the searcher in control of the search, on the other hand the searcher now has to know which word to use to express his information need when looking for documents, and every so often he has to know how to use the tool that performs such search.
With the increasing availability of electronic text and with the searcher becoming the user of an information accessing system, it became necessary to develop systems that were both easy to use and effective. The birth of the World Wide Web has magnified enormously this problem, as everybody knows.
So, in over 50 years researchers in IR have developed and evaluated a bewildering array of techniques for indexing and retrieving information. These techniques have slowly matured and improved through refinement, rather than there having been one or a small number of really significant break throughs.
The purpose of the Third European Summer School in Information (ESSIR'2000), held in Varenna (Italy) at Villa Monastero in September 2000, was to pass on to younger generations of researchers the expertise and knowledge acquired by some of the best European experts in IR and related areas. There is a widely perceived need of new developments in IR and, like in many other fields of research, significant breakthroughs are more likely to come from younger minds than from experienced researchers. Indeed, there was a very young level of participation at ESSIR'2000. As organisers, we have very happy of this. We hope that more advances and, perhaps, some breakthrough, will come to IR from this young international crowd that for a week attended the school and strolled along the shores of Lake Como. Here is a short report of that event.
The Third European Summer School in Information (ESSIR'2000) was held at Villa Monastero in Varenna (Italy) in September 2000 [3]. It is part of a series of ESSIRs that began in 1990, the first one was organised by the University of Padova (by Maristella Agosti) and was held in Bressanone, Italy in 1990. The second ESSIR was organised by the University of Glasgow (by Keith van Rijsbergen) and held in Glasgow in 1995, in the context of the IR Festival which consisted of ESSIR '95 [5], a IRIDES workshop [6], and the final MIRO workshop [7]. ESSIR'2000 was jointly organised by:
The administrative support was provided by Milano Ricerche, a consortium of industries, research institutions (CNR among them) and the University of Milano, whose purpose is to provide administrative and technical support to research and development activities of its members.
The scope of ESSIR'2000 was to give to its participants a grounding in the core subjects of IR, which included methods and techniques for designing and developing IR systems, Web search engines, and tools for information storing and querying in Digital Libraries. To reach this scope, the program of the ESSIR 2000 was been organised over the following lectures: an introduction to IR, fundamental IR models, evaluation in IR, multi-lingual information access, multimedia IR (audio image, and video), digital libraries, IR and users, uncertainty and logic in IR, modelling vagueness in IR, IR on the Web, and IR and structured documents.
The lecturers of the school were leading European researchers (with only one extra-European exception). Their course subjects strongly reflected the research work for which they are all well known.
ESSIR'2000 was intended for researchers starting out in IR, for industrialists who wish to know more about this increasingly important topic and for people working on topics related to management of information on the Internet. The proceedings, (that were distributed at the school in draft and that will be available in final form shortly [4]) contains 12 chapters written by the school's lecturers detailing the state of the art in IR and related areas. They contain experience distilled from many years of work.
The programme of the school was very dense, comprising of 11 lectures, divided in basic (6 lectures) and advanced (5 lectures). Here is briey what was taught and who did it. More information can be found on teh Web site [3], [4].
Keith van Rijsbergen (University of Glasgow) gave the opening lecture by introducing some underlying concepts and ideas essential for understanding IR research and techniques. He also highlighted some related hot areas of research, emphasising the role of IR in each.
Norbert Fuhr (University of Dortmund) lectured on main mathematical models of IR. His lecture gave the theoretical basis for representing the informative content of documents and for estimating the relevance of a document to a query.
Paraic Sheridan (TextWise) and Carol Peters (IEI-CNR) presented, in a nicely concerted way, the issues and proposed solutions to multi-lingual information access in digital archives.
Stephen Robertson (Microsoft) addressed the topic of evaluation.
Alan Smeaton (Dublin City University) and John Eakins (University of Northumbria) addressed issues and techniques related to indexing, browsing and searching multimedia information (audio, image, or digital video).
The lecture by Ingeborg Solvberg (Norwegian University of Science and Technology) covered the basics and the challenges of digital libraries.
The lecture by Peter Ingwersen (Danish Royal School of Library and Information Science), the First on advanced topics, concentrated on users issues and usability of interactive IR.
Fabio Crestani (University of Strathclyde) and Mounia Lalmas (University of London) addressed the use of logic and uncertainty theories in IR.
Gabriella Pasi and Gloria Bordogna (both from ITIM-CNR) presented the area of research in that aims at modelling the vagueness and imprecision involved in the IR process.
Maristella Agosti and Massimo Melucci (both at the University of Padova) addressed the use of IR techniques on the Web.
Finally, Yves Chiaramella (University of Grenoble) addressed the issues related to indexing and retrieval of structured documents.
The school was a success. Not just in the words of the organisers (whom you would expect to claim so!), but and most importantly by the judgement of the participants. ESSIR'2000 was a success not just for the quality of the quality of the lectures, the authority of the the lecturers, and the beautiful surroundings, it was a success because it was informal and interactive. For the best part of a week more that 60 participants and 12 lecturers exchanged ideas and inspirations on where IR is at and where it should go to. Many (not only the school participants, but some of the lecturers too) went home with renewed encouragements and motivations.
Not everything ran smoothly (as Murphy's law teaches us). There were tense moments when Fuhr's demos did not start and when in the middle of Robertson's lecture the electricity went off. There was also terror in the eyes of the participants when, after the official dinner on the other side of Lake Como (in Belaggio), the boat chartered to take everybody back to Varenna did not show up on time. The thought of a long cold swim after such a wonderfully filling meal put the life of the organisers in serious danger. We are sure other faults can be found in the organisation of the event. Nevertheless, the vast majority of participants went home very happy with their experience. That is not only the impression that we gathered but it is also the feedback that we received and are still receiving.
Finally, from the point of view of the organisers, it was very hard work. It took months to organise the event and although it is now over, we are still working for it. Nevertheless it was a useful and enjoyable experience from which we have learned a lot.
So far, ESSIRs have been held at a five-year interval. We hope this interval will be considerably reduced in the future. The large participation of young researchers to ESSIR'2000 shows that there is a very active interest in the field and one that is likely to grow even more in the future. We are ready to pass the torch to the next organisers, to whom we will be happy provide our experience in the organisation of such an event.
Indeed, there was a very young participation at the Third European Summer School in Information (ESSIR'2000), that was held in Varenna (Italy) by the Lake Como, in September 2000. Probably a much younger participation than other events held there. As organisers, we are very happy about this. We hope that more advances and, perhaps, some breakthroughs, will come to IR from this young international crowd that for a week, thoughtfully but joyfully, strolled along the shores of Lake Como.
The main financial support for ESSIR'2000 was provide by the registration fees and by the the Special Interest Network on Information Retrieval of the Council of European Professional Informatics Societies (CEPIS-IR). In particular, CEPIS-IR made available a number of grants for young students and researchers to attend the school.
Other financial support was provide by the following sponsors: Microsoft Italia, Mondadori, Oracle, Sharp Laboratories of Europe, and 3-D Informatica.
Finally, the following institutions provided general support: the Gruppo Specialistico Tecnologie e Applicazioni Informatiche (AEI) and the Convention of National Societies of Electrical Engineers of Europe (EUREL).
Maristella Agosti
Dipartimento di Elettronica e Informatica
Universita' di Padova
Padova, Italy
URL: <http://www.dei.unipd.it/>
Email: agosti@dei.unipd.it
Fabio Crestani
Department of Computer Science
University of Strathclyde
Glasgow, G1 1XH, Scotland
URL: <http://www.cs.strath.ac.uk/>
Email: fabioc@cs.strath.ac.uk
Gabriella Pasi
Istituto Technologie Informatiche Multimediali
Consiglio Nazionale delle Ricerche
Milano, Italy
URL: <http://www.itim.mi.cnr.it/>
Email: gabriella.pasi@itim.mi.cnr.it
For citation purposes:
Maristella Agosti, Fabio Crestani and Gabriella Pasi, "ESSIR'2000: Information Retrieval by the Lake", Exploit Interactive, issue
7, 2nd October 2000
URL: <http://www.exploit-lib.org/issue7/essir/>
[HTML Validation] - [Accessibility check]
|
Issue Home | Editorial | Features | Regular Columns | News and Events | Et cetera | ||
|
| ||
| Go to Top |
A UKOLN Service. Contact Us. Copyright © 1999 |
Last Updated: 2nd October 2000 |