Focused Retrieval and Evaluation

Welcome to the proceedings of the 8th Workshop of the Initiative for the Eva- ation of XML Retrieval (INEX)! Now in its eighth year, INEX is an established evaluation forum for XML information retrieval (IR), with over 100 organi- tions worldwide registered and over 50 groups participating actively in at least one of the tracks. INEX aims to provide an infrastructure, in the form of a large structured test collection and appropriate scoring methods, for the evaluation of focused retrieval systems. XML IR plays an increasingly important role in many information access systems (e.g., digitallibraries,Web, intranet) where content is a mixture oftext, multimedia, and metadata, formatted according to the adopted W3C standard for information repositories, the so-called eXtensible Markup Language (XML). The ultimate goal of such systems is to provide the right content to their e- users. However, while many of today s information access systems still treat documents as single large (text) blocks, XML o?ers the opportunity to exploit the internal structure of documents in order to allow for more precise access, thus providing more speci?c answers to user requests. Providing e?ective access to XML-based content is therefore a key issue for the success of these systems. INEX2009wasanexcitingyearforINEXinwhichanewcollectionwasint- duced that is again based on Wikipedia but is more than four times larger, with longerarticlesandadditionalsemanticannotation.Intotal,eightresearchtracks were included, which studied di?erent aspects of focused information access: Ad Hoc Track investigatedthee?ectivenessofXML-IRandPassageRetrieval for four ad hoc retrievaltasks:Thorough,Focused, Relevant in Context, and Best in Context.

Up-to-date results Fast track conference proceedings State-of-the-art report

Klappentext

This book constitutes the thoroughly refereed post-workshop proceedings of the 8th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2009, held in Brisbane, Australia, in December 2009. The aim of the INEX 2009 workshop was to bring together researchers in the field of XML IR who participated in the INEX 2009 campaign. During the past year, participating organizations contributed to the building of large-scale XML test collections by creating topics, performing retrieval runs and providing relevance assessments. The workshop concluded the results of this effort, summarized and addressed issues encountered, and devised a work plan for the future evaluation of XML retrieval systems. The 42 full papers presented together with 3 invited papers were carefully reviewed and selected from 49 submissions. They have been divided into sections according to the eight tracks of the workshop, investigating various aspects of XML retrieval, from book search to entity

Inhalt
Invited.- Is There Something Quantum-Like about the Human Mental Lexicon?.- Supporting for Real-World Tasks: Producing Summaries of Scientific Articles Tailored to the Citation Context.- Semantic Document Processing Using Wikipedia as a Knowledge Base.- Ad Hoc Track.- Overview of the INEX 2009 Ad Hoc Track.- Analysis of the INEX 2009 Ad Hoc Track Results.- ENSM-SE at INEX 2009 : Scoring with Proximity and Semantic Tag Information.- LIP6 at INEX'09: OWPC for Ad Hoc Track.- A Methodology for Producing Improved Focused Elements.- ListBM: A Learning-to-Rank Method for XML Keyword Search.- UJM at INEX 2009 Ad Hoc Track.- Language Models for XML Element Retrieval.- Use of Language Model, Phrases and Wikipedia Forward Links for INEX 2009.- Parameter Tuning in Pivoted Normalization for XML Retrieval: ISI@INEX09 Adhoc Focused Task.- Combining Language Models with NLP and Interactive Query Expansion.- Exploiting Semantic Tags in XML Retrieval.- Book Track.- Overview of the INEX 2009 Book Track.- XRCE Participation to the 2009 Book Structure Task.- The Book Structure Extraction Competition with the Resurgence Software at Caen University.- Ranking and Fusion Approaches for XML Book Retrieval.- OUC's Participation in the 2009 INEX Book Track.- Efficiency Track.- Overview of the INEX 2009 Efficiency Track.- Index Tuning for Efficient Proximity-Enhanced Query Processing.- TopX 2.0 at the INEX 2009 Ad-Hoc and Efficiency Tracks.- Fast and Effective Focused Retrieval.- Achieving High Precisions with Peer-to-Peer Is Possible!.- Entity Ranking Track.- Overview of the INEX 2009 Entity Ranking Track.- Combining Term-Based and Category-Based Representations for Entity Search.- Focused Search in Books and Wikipedia: Categories, Links and Relevance Feedback.- A Recursive Approach to EntityRanking and List Completion Using Entity Determining Terms, Qualifiers and Prominent n-Grams.- Interactive Track.- Overview of the INEX 2009 Interactive Track.- Link the Wiki Track.- Overview of the INEX 2009 Link the Wiki Track.- An Exploration of Learning to Link with Wikipedia: Features, Methods and Training Collection.- University of Waterloo at INEX 2009: Ad Hoc, Book, Entity Ranking, and Link-the-Wiki Tracks.- A Machine Learning Approach to Link Prediction for Interlinked Documents.- Question Answering Track.- Overview of the 2009 QA Track: Towards a Common Task for QA, Focused IR and Automatic Summarization Systems.- XML Mining Track.- Overview of the INEX 2009 XML Mining Track: Clustering and Classification of XML Documents.- Exploiting Index Pruning Methods for Clustering XML Collections.- Multi-label Wikipedia Classification with Textual and Link Features.- Link-Based Text Classification Using Bayesian Networks.- Clustering with Random Indexing K-tree and XML Structure.- Utilising Semantic Tags in XML Clustering.- UJM at INEX 2009 XML Mining Track.- BUAP: Performance of K-Star at the INEX'09 Clustering Task.- Extended VSM for XML Document Classification Using Frequent Subtrees.- Supervised Encoding of Graph-of-Graphs for Classification and Regression Problems.