CLEF 2010 had two different types of labs: benchmarking evaluations for specific information access problem(s), encompassing activities similar to previous CLEF campaign tracks, and workshops for exploration of specific issues in information access. A call was made for proposals for both types of labs, with 9 proposals being submitted in late October of 2009. Five of these were accepted as benchmarking activites, and two were accepted as exploration workshops. Each submitted proposal was reviewed by the CLEF2010 lab organizing committee, with final decisions of acceptance and length of labs determined by the following criteria.
– PAN, a lab on uncovering plagiarism, authorship, and social software misuse, ran at CLEF for the first time, following three previous workshops at other conferences. It was sponsored by Yahoo Research and had two tasks, namely the detection of plagiarism and the detection of Wikipedia vandalism. – QA@CLEF 2010 - ResPubliQA was the eighth year for multilingual question answering in CLEF. Similar to the ResPublicQA version in CLEF2009, the lab used the Europarl Corpus, and had seven monolingual tasks for English, French, German, Italian, Portuguese, Spanish and Romanian. – WePS (Web People Search) focused on person name ambiguity and person attribute extraction on Web pages and on Online Reputation Management (ORM) for organizations, again dealing with the problem of ambiguity for organization names and the relevance of Web data for reputation management purposes. This was the lab’s first year at CLEF, following two previous workshops at other conferences.
These two exploration workshops ran as labs in CLEF 2010.
– CriES addressed the problem of multi-lingual expert search in social media environments. The main topics were multi-lingual expert retrieval methods, social media analysis with respect to expert search, selection of datasets and evaluation of expert search results. Papers reporting on experiments or proposals for possible benchmarking activities were invited. – LogCLEF investigated the analysis and classification of queries in order to understand search behavior in multilingual contexts and ultimately to improve search systems. The different log sets were used, the The European Library (TEL) logs, and the Deutscher Bildungsserver (DBS) logs, a quality controlled internet directory for educational resources. Participants were invited to investigate a variety of questions with the end goal of defining a benchmarking task for follow-on labs. We would also like to thank all of the people involved in making these labs happen; they are the heart of CLEF.