Corrado Boscarino

DIR 2011, the 11th Dutch-Belgian Information Retrieval Workshop, was organized by the Information and Language Processing group (ILPS) of the University of Amsterdam in collaboration with the Centrum Wiskunde en Informatica (CWI). Two types of submissions were accepted for the workshop: research papers describing original research, compressed contributions presenting a summary of previously published work, and demonstrations. There were many people who helped organize DIR 2011, making it a success. We would like to thank them all. In particular, we are gratefull to our keynote speakers, Nick Belkin (Rutgers University) and Gabriella Kazai (Microsoft Research). Relevance has been the classic criterion for evaluation of the effectiveness of information retrieval (IR) systems since the earliest days of IR system evaluation. This criterion has been understood as the ability of an IR system to recognize documents relevant to a person's " information need " , and understood ...

The 11th edition of the annual Dutch-Belgian Information Retrieval workshop (DIR 2011) took place on February 4 in Amsterdam. It was organized by the University of Amsterdam and the Centrum Wiskunde & Informatica. The focus of this year's workshop was on interaction, with the goal of facilitating and increasing interaction, especially within the local research community, and between industry and academia. The scientific program included demos, research papers, and compressed contributions. The keynotes by Nick Belkin and Gabriella Kazai provided intriguing outlooks on the future of IR evaluation.

htmlabstractWe argue that query expansion (QE) based on the full ses- sion improves the overall search experience provided that we know how to adapt the QE weighting schema to a user's search proficiency. We propose a strategy to predict search ability from session parameters. Us- ing an exponential model and these metrics we set user dependent QE coefficients. We evaluate this approach on TREC 2011 session track data.

We participated in two tracks: Knowledge Base Acceleration (KBA) Track and Session Track. In the KBA track, we focused on experi- menting with different approaches as it is the first time the track is launched. We experimented with supervised and unsupervised re- trieval models. Our supervised approach models include language models and a string-learning system. Our unsupervised approaches include using: 1)DBpedia labels and 2) Google-Cross-Lingual Dic- tionary (GCLD). While the approach that uses GCLD targets the central and relvant bins, all the rest target the central bin. The GCLD and the string-learning system have outperformed the oth- ers in their respective targeted bins. The goal of the Session track submission is to evaluate whether and how a logic framework for representing user interactions with an IR system can be used for improving the approximation of the relevant term distribution that another system that is supposed to have access to the session infor- mation will t...

La presente invention a trait a un codeur et un decodeur audio et a des procedes pour le codage et decodage audio. Dans le codeur un signal audio est divise en une partie de signal anechoique et une information concernant un champ reverbere associe au signal audio, de preference par une representation n'utilisant que tres peu de parametres tel que le temps de reverberation et l'amplitude de reverberation. Le signal anechoique est ensuite code au moyen d'un codeur-decodeur audio. Au niveau du decodeur la partie de signal anechoique restituee est transformee en un signal audio sensiblement d'origine par l'application de reverberation selon l'information concernant le champ reverbere, de preference par convolution avec une reponse impulsionnelle de chambre generee en fonction de l'information de champ reverbere. Selon l'invention, le codeur-decodeur audio concerne n'a seulement besoin d'etre capable de codage de signaux audio anechoiques, permett...

Abstract. We evaluate the use of clickthrough information as implicit relevance feedback in sessions. We employ records of user interactions with a commercial news picture portal: issued queries, clicked images, and purchased content. Our study investigates how much of a session's search history (if any) should be used in a feedback loop. We assess the benefit of using clicked data as positive tokens of relevance to the task of estimating the probability of an image to be purchased. We find that a short history of past queries helps improve ...

We investigate how a user-centred design to search can improve the support of user tasks specific to journalism. Illustrated by example information needs, sampled from our own exploration of the New York Times annotated corpus, we demonstrate how domain specific notions rooted in a field theory of journalism can be transformed into effective search strategies. We present a method for search-context aware classification of authorities, witnesses, reporters and columnists. A first search strategy supports the journalistic task of ...

There is a wide agreement that a user centred approach to IR applications design outperforms system centred ones. A classic understanding of this design approach, and specifically of its underlying notion of context, appeared however insufficient to explain the results of a pilot experiment. We recognise the importance of context, but we define context differently by means of a domain theory: a conceptualisation of the domain at hand, preferably developed within the same community which users belong to.

Abstract. A mismatch between different event spaces has been used to argue against rank equivalence of classic probabilistic models of information retrieval and language models. We question the effectiveness of this strategy and we argue that a convincing solution should be sought in a correct procedure to design adequate priors for probabilistic reasoning. Acknowledging our solution of the event space issue invites to rethink the relation between probabilistic models, statistics and logic in the context of IR.

The so called logic and probabilistic views on IR can be reconciled by a unifying framework for IIR. I present a proposal for a PhD research according to a multidisciplinary perspective and I discuss some of its consequences for IR as a discipline.