Advances in XML Information Retrieval and Evaluation: 4th by Norbert Fuhr, Mounia Lalmas, Saadia Malik, Gabriella Kazai

By Norbert Fuhr, Mounia Lalmas, Saadia Malik, Gabriella Kazai

Content-oriented XML retrieval has been receiving expanding curiosity as a result of the common use of eXtensible Markup Language (XML), that's turning into a customary rfile structure on the net, in electronic libraries,and publishing. via exploiting the enriched resource of syntactic and semantic info that XML markup offers, XML info retrieval (IR) platforms objective to enforce a extra centred retrieval approach and go back rfile parts, so-called XML parts – rather than whole files – based on a consumer question. This concentrated retrieval process is of specific bene?t for collections containing lengthy records or records protecting a large choice of themes (e.g., books, person manuals, felony records, etc.), the place clients’ e?ort to find correct content material will be diminished via directing them to the main correct components of the records. enforcing this, extra concentrated, retrieval paradigm implies that an XML IR approach wishes not just to ?nd proper info within the XML records, however it additionally has to figure out the ideal point of granularity to be lower back to the person. moreover, the relevance of a retrieved part will be depending on assembly either content material and structural question conditions.

Show description

Read or Download Advances in XML Information Retrieval and Evaluation: 4th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2005, Dagstuhl Castle, Germany, November 28-30, 2005. Revised Selected Papers PDF

Best storage & retrieval books

Introduction to Shannon Sampling and Interpolation Theory

Regaining unique signs reworked from analog to electronic platforms or assessing details misplaced within the method are the elemental concerns addressed through sampling and interpolation conception. This research makes an attempt to appreciate, generalize and expand the cardinal sequence of Shannon sampling idea.

Semantic Web Services: Theory, Tools and Applications

The Semantic net proposes the mark-up of content material on the internet utilizing formal ontologies that constitution underlying information for the aim of complete and conveyable laptop realizing. Semantic net companies: concept, instruments and functions brings contributions from researchers, scientists from either and academia, and representatives from assorted groups to review, comprehend, and discover the idea, instruments, and purposes of the semantic internet.

A Unified Framework for Video Summarization, Browsing & Retrieval: with Applications to Consumer and Surveillance Video

Huge volumes of video content material can basically be simply accessed via speedy searching and retrieval options. developing a video desk of contents (ToC) and video highlights to permit finish clients to sift via all this information and locate what they need, after they wish are crucial. This reference places forth a unified framework to combine those features aiding effective searching and retrieval of video content material.

Yahoo! to the Max: An Extreme Searcher Guide

Yahoo! has a variety of how one can locate details, converse, make investments, store, and promote, and this publication offers an outline of the preferred internet portal. information on internet looking, discovering and customizing information, and utilizing and developing chat groups are integrated, in addition to details on utilizing Yahoo!

Additional resources for Advances in XML Information Retrieval and Evaluation: 4th International Workshop of the Initiative for the Evaluation of XML Retrieval, INEX 2005, Dagstuhl Castle, Germany, November 28-30, 2005. Revised Selected Papers

Example text

E1) is 1; (E2) is resp. 9 2 − 2 ) = 4 for lists A, B, C and D. 8 2 , 4 and 4 . 40 B. Piwowarski Recall 3 (level 1). (E1) is 2; (E2) is resp. 9 4 for lists A, B, C and D. Precisions are 1, 1, 2 and 2 . Using the same technique that in the previous section, we now combine the precisions for these two sets of users. Table 4. Precision-recall for the list A-D (Fetch&Browse). The precision for the four lists and four recall intervals are shown. The line “correspondence” shows what is the number of ideal elements the user wants to see if, among the elements in the ideal set, (1) he considers that only elements with an exhaustivity 2 are ideal (2) he considers that elements with an exhaustivity at least 1 are ideal.

This model has two interesting implications in XML IR: The notion of ideal list. The ideal list is used both by EPRUM and xCG metrics to compute respectively the ideal precision/recall and effort/gain curves. While in xCG the notion of ideal list and ideal set of elements are the same, this is not the case in EPRUM: The ideal set of elements is, as in xCG, the set of elements that would satisfy users with the same query need. Contrarily to xCG, the ideal list of EPRUM is not always the set of ideal elements and depends on the specific user model.

12 , 12 , 12 × ( 12 − 0) = 14 , and 14 for lists A, B, C and D. Precisions is 1, 1, 12 and 12 . 2. If the user is satisfied with an element of exhaustivity at least 1 (25 % of the users): 1 Recall 1 (level 1/3). (E1) is 1; (E2) is resp. 9 2 ) = 4 for lists A, B, C and D. Precision is 1, 1, 1, and 4 . Recall 2 (level 2/3). (E1) is 1; (E2) is resp. 9 2 − 2 ) = 4 for lists A, B, C and D. 8 2 , 4 and 4 . 40 B. Piwowarski Recall 3 (level 1). (E1) is 2; (E2) is resp. 9 4 for lists A, B, C and D. Precisions are 1, 1, 2 and 2 .

Download PDF sample

Rated 4.84 of 5 – based on 16 votes