Advances in focused retrieval : 7th international workshop of the Initiative for the Evaluation of Xml Retrieval, INEX 2008, Dagstuhl Castle, Germany, December 15-18, 2008 : revised and selected papers

This book constitutes the thoroughly refereed proceedings of the 7th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2008, held at Dagstuhl Castle, Germany, in December 2008. The aim of the INEX 2008 workshop was to bring together researchers who participated in the INEX 2008 campaign. Over the year leading up to the event, participating organizations contributed to the building of a large-scale XML test collection by creating topics, performing retrieval runs, and providing relevance assessments. The workshop concluded the results of this large-scale effort, summarized and addressed the issues encountered, and devised a work plan for the future evaluation of XML retrieval systems. The 49 papers included in this volume report the final results of INEX 2008. They have been divided into sections according to the seven tracks of the workshop, investigating various aspects of XML retrieval, from book search to entity ranking, including interaction aspects. Front Matter....Pages - Overview of the INEX 2008 Ad Hoc Track....Pages 1-28 Experiments with Proximity-Aware Scoring for XML Retrieval at INEX 2008....Pages 29-32 Finding Good Elements for Focused Retrieval....Pages 33-38 New Utility Models for the Garnata Information Retrieval System at INEX’08....Pages 39-45 UJM at INEX 2008: Pre-impacting of Tags Weights....Pages 46-53 Use of Multiword Terms and Query Expansion for Interactive Information Retrieval....Pages 54-64 Enhancing Keyword Search with a Keyphrase Index....Pages 65-70 CADIAL Search Engine at INEX....Pages 71-78 Indian Statistical Institute at INEX 2008 Adhoc Track....Pages 79-86 Using Collectionlinks and Documents as Context for INEX 2008....Pages 87-96 SPIRIX: A Peer-to-Peer Search Engine for XML-Retrieval....Pages 97-105 Overview of the INEX 2008 Book Track....Pages 106-123 XRCE Participation to the Book Structure Task....Pages 124-131 University of Waterloo at INEX 2008: Adhoc, Book, and Link-the-Wiki Tracks....Pages 132-139 The Impact of Document Level Ranking on Focused Retrieval....Pages 140-151 Adhoc and Book XML Retrieval with Cheshire....Pages 152-163 Book Layout Analysis: TOC Structure Extraction Engine....Pages 164-171 The Impact of Query Length and Document Length on Book Search Effectiveness....Pages 172-178 Overview of the INEX 2008 Efficiency Track....Pages 179-191 Exploiting User Navigation to Improve Focused Retrieval....Pages 192-206 Efficient XML and Entity Retrieval with PF/Tijah: CWI and University of Twente at INEX’08....Pages 207-217 Pseudo Relevance Feedback Using Fast XML Retrieval....Pages 218-223 TopX 2.0 at the INEX 2008 Efficiency Track....Pages 224-236 Aiming for Efficiency by Detecting Structural Similarity....Pages 237-242 Overview of the INEX 2008 Entity Ranking Track....Pages 243-252 L3S at INEX 2008: Retrieving Entities Using Structured Information....Pages 253-263 Adapting Language Modeling Methods for Expert Search to Rank Wikipedia Entities....Pages 264-272 Finding Entities in Wikipedia Using Links and Categories....Pages 273-279 Topic Difficulty Prediction in Entity Ranking....Pages 280-291 A Generative Language Modeling Approach for Ranking Entities....Pages 292-299 Overview of the INEX 2008 Interactive Track....Pages 300-313 Overview of the INEX 2008 Link the Wiki Track....Pages 314-325 Link-the-Wiki: Performance Evaluation Based on Frequent Phrases....Pages 326-336 CMIC@INEX 2008: Link-the-Wiki Track....Pages 337-342 Stealing Anchors to Link the Wiki....Pages 343-353 Context Based Wikipedia Linking....Pages 354-365 Link Detection with Wikipedia....Pages 366-373 Wikisearching and Wikilinking....Pages 374-388 CSIR at INEX 2008 Link-the-Wiki Track....Pages 389-394 A Content-Based Link Detection Approach Using the Vector Space Model....Pages 395-400 Overview of the INEX 2008 XML Mining Track....Pages 401-411 Semi-supervised Categorization of Wikipedia Collection by Label Expansion....Pages 412-419 Document Clustering with K-tree....Pages 420-431 Using Links to Classify Wikipedia Pages....Pages 432-435 Clustering XML Documents Using Frequent Subtrees....Pages 436-445 UJM at INEX 2008 XML Mining Track....Pages 446-452 Probabilistic Methods for Link-Based Classification at INEX 2008....Pages 453-459 Utilizing the Structure and Content Information for XML Document Clustering....Pages 460-468 Self Organizing Maps for the Clustering of Large Sets of Labeled Graphs....Pages 469-481 Back Matter....Pages - I write with pleasurethis forewordto the proceedings of the 7th workshopof the Initiative for the Evaluation of XML Retrieval (INEX). The increased adoption of XML as the standard for representing a document structure has led to the development of retrieval systems that are aimed at e?ectively accessing XML documents. Providing e?ective access to large collections of XML documents is therefore a key issue for the success of these systems. INEX aims to provide the necessary methodological means and worldwide infrastructures for evaluating how good XML retrieval systems are. Since its launch in 2002, INEX has grown both in terms of number of p- ticipants and its coverage of the investigated retrieval tasks and scenarios. In 2002, INEX started with 49 registered participating organizations, whereas this number was more than 100 for 2008. In 2002, there was one main track, c- cerned with the ad hoc retrieval task, whereas in 2008, seven tracks in addition to the main ad hoc track were investigated, looking at various aspects of XML retrieval, from book search to entity ranking, including interaction aspects.