When the objective of an information retrieval task is to return a nugget rather than a document, query terms that exist in a document often will not be used in the most relevant nugget in the document for the query. In this thesis a new method of query expansion is proposed based on the Wikipedia link structure surrounding the most relevant articles selected either automatically or by human assessors for the query. Evaluated with the Nuggeteer automatic scoring software, which we show to have a high correlation with human assessor scores for the ciQA 2006 topics, an increase in the F-scores is found from the TREC Complex Interactive Question Answering task when integrating this expansion into an already high-performing baseline system. In addition, the method for finding synonyms using Wikipedia is evaluated using more common synonym detection tasks.

Thesis (M.Sc.)--York University, 2006. Graduate Programme in Higher Education. / Typescript. Includes bibliographical references (leaves 93-96). Also available on the Internet. MODE OF ACCESS via web browser by entering the following URL: http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&res_dat=xri:pqdiss&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&rft_dat=xri:pqdiss:MR29598

Thesis (Ph. D.)--University of Hong Kong, 2001. / Includes bibliographical references (leaves 94-98).

Thesis (M.S.)--Brigham Young University. Dept. of Computer Science, 2003. / Includes bibliographical references (p. 65-69).

This study investigates the use of criteria to assess relevant, partially relevant and not relevant documents. Each study participant identified passages within 20 document representations that were used in making relevance judgments, judged each document representation as a whole to be relevant, partially relevant or not relevant to their information need, and explained their decisions in an interview. Analysis revealed 29 criteria, discussed positively and negatively, used by the participants when selecting passages that contributed or detracted from a document's relevance. These criteria can be grouped into 6 categories: author, abstract, content, full text, journal or publisher and personal. Results indicate that multiple criteria are used when making relevant, partially relevant and not relevant judgments. Additionally, most criteria can have both a positive or negative contribution to the relevance of a document. The criteria most frequently mentioned by study participants in this study was content, followed by criteria concerning the full text document. These findings may have implications for relevance feedback in information retrieval systems, suggesting that users give relevance feedback using multiple criteria and indicate positive and negative criteria contributions. Systems designers may want to focus on supporting content criteria followed by full text criteria as this may provide the greatest cost benefit.

In this paper we study three OPAC interfaces of K.U.Leuven University Library. All three interfaces have been on-line between 2002 and 2005. The characteristics of these OPACs (search fields, labeling of the fields, search facilities, searching vs. browsing, basic vs. advanced search) are systematically examined. Special attention is given to subject search and display, and more specifically to Universal Decimal Classification (UDC) and Medical Subject Headings (MeSH). This comparison is particularly interesting because the indexing and classification tools (UDC, Library of Congress Subject Headings, MeSH) remain the same in all three OPACs. However, the way these tools are presented for subject retrieval in the respective OPACs undergoes an interesting evolution. It is demonstrated that subject search and display in library catalogs can be largely improved by investing in optimal use of existing tools. We also stress the importance of user-oriented OPAC design.

This is a 27-slide presentation made at ISKO 2004 (International Society for Knowledge Organization) in London, UK. Recent projects in subject indexing and classification at K.U.Leuven University Library (Belgium) aim to give new impulses to knowledge organization within the institution. While in recent years a lot of attention was given, and with good reason, to the technical and administrative integration of e-sources, less energy was invested in organizing the content of traditional and electronic collections. Nevertheless, presenting information sources in a content-structured way remains a core task of our University Library. This paper focuses on some experiments with subject search interfaces at K.U.Leuven University Library and situates them in a new policy for knowledge organization, which tries to find a balance between local and overall needs.

Name Search is an important search function in various types of information retrieval systems, such as online library catalogs and electronic yellow pages. It is also difficult due to the high degree of fuzziness required in matching name variants. Previous approaches to name search systems use ad hoc combinations of search heuristics. This paper first discusses two approaches to name modelingâ the natural language processing (NLP) and the information retrieval (IR) modelsâ and proposes a hybrid approach. The approach demonstrates a critical combination of complementary NLP and IR features that produces more effective fuzzy name matching. Two principles, position-as-attribute and position-transitionlikelihood, are introduced as the principles for integrating the advantageous aspects of both approaches. They have been implemented in an NLP- and IR- hybrid model system called Friendly Name Search (FNS) for real world applications in multilingual directory searches on the Singapore Yellow pages website.

ICT for Library and Information Professionals: A Training Package for Developing Countries (ICTLIP) edited by Andrew Large, McGill University, and published by the Information and Informatics Unit, UNESCO Bangkok, Thailand, consists of 6 modules, namely, Module 1: Introduction to information and communication technologies; Module 2: Introduction to integrated library systems; Module 3: Information seeking in an electronic environment; Module 4: Database design, and information storage and retrieval; Module 5: The Internet as an information resource; and Module 6: Web page concept and design: getting a web site up and running. This is Module 4 Database design, and information storage and retrieval (Teacher's Guide) by Deepali Talagala. Module 4 concerns the concepts of database designing using WinISIS, and the storage and retrieval of information. By the end of the Module, students should: * Be able to use the basic features of WinISIS *Be able to identify the principles of database designing *Be able to design and use a sample database *Understand the technological features *Know how to use

Στην παρούσα διπλωματική εργασία παρουσιάζονται μέθοδοι και εργαλεία προεπεξεργασίας και ανάκτησης πληροφορίας συλλογών κειμένου. Το σύνολο των μεθόδων αυτών στηρίζονται στο μοντέλο Διανυσματικού Χώρου και χρησιμοποιούν στον πυρήνα τους αλγορίθμους από την περιοχή της Υπολογιστικής Γραμμικής Άλγεβρας. Πιο συγκεκριμένα, στην παρούσα εργασία παρουσιάζεται αρχικά το εργαλείο λογισμικού Text to Matrix Generator (TMG), για την κατασκευή μητρώων όρων-κειμένων από συλλογές κειμένου, που είναι γραμμένο εξ / We consider algorithms and tools for text preprocessing and information retrieval. All these algorithms are based on the Vector Space Model and use as kernel Computational Linear Algebra algorithms. In particular, we present the text preprocessing software tool Text to Matrix Generator (TMG), that can be used for the construction of term-document matrices from text collections and is written entirely in MATLAB. TMG can be used for the application of a wide range of text preprocessing techniques, such as stopword removal and stemming. Furthermore, TMG can be used for the construction of query vectors using the collection