Search (13 results, page 1 of 1)

0.044312872 = product of:
  0.088625744 = sum of:
    0.088625744 = sum of:
      0.04620442 = weight(_text_:systems in 2717) [ClassicSimilarity], result of:
        0.04620442 = score(doc=2717,freq=4.0), product of:
          0.16037072 = queryWeight, product of:
            3.0731742 = idf(docFreq=5561, maxDocs=44218)
            0.052184064 = queryNorm
          0.28811008 = fieldWeight in 2717, product of:
            2.0 = tf(freq=4.0), with freq of:
              4.0 = termFreq=4.0
            3.0731742 = idf(docFreq=5561, maxDocs=44218)
            0.046875 = fieldNorm(doc=2717)
      0.042421322 = weight(_text_:22 in 2717) [ClassicSimilarity], result of:
        0.042421322 = score(doc=2717,freq=2.0), product of:
          0.1827397 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.052184064 = queryNorm
          0.23214069 = fieldWeight in 2717, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.046875 = fieldNorm(doc=2717)
  0.5 = coord(1/2)

Abstract

Previous work an search-engine design has indicated that information-seekers may benefit from being given the opportunity to exploit multiple sources of evidence of document relatedness. Few existing systems, however, give users more than minimal control over the selections that may be made among methods of exploitation. By applying the methods of "document network analysis" (DNA), a unifying, graph-theoretic model of content-, collaboration-, and context-based systems (CCC) may be developed in which the nature of the similarities between types of document relatedness and document ranking are clarified. The usefulness of the approach to system design suggested by this model may be tested by constructing and evaluating a prototype system (UCXtra) that allows searchers to maintain control over the multiple ways in which document collections may be ranked and re-ranked.

Date

11. 9.2004 17:32:22

0.02161014 = product of:
  0.04322028 = sum of:
    0.04322028 = product of:
      0.08644056 = sum of:
        0.08644056 = weight(_text_:systems in 2044) [ClassicSimilarity], result of:
          0.08644056 = score(doc=2044,freq=14.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.5390046 = fieldWeight in 2044, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.046875 = fieldNorm(doc=2044)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

We have recently completed a survey of the operational use of hypertext-based information systems in academic, public and special libraries in the UK. A literatur search, questionnaire and both telephone and face-to-face interviews demonstrate that the principle application of hypertext systems is for the implementation of public-access point-of-information systems, which provide guidance to the users of local information resources. In this paper, we describe the principle issuse relating to the design and usage of these systems that were raised in the interviews and that we experienced when using the systems for ourselves. We then present a set of technical recommendations with the intention of helping the developers of future systems, with special attention being given to the need to develop effective methods for system evaluation

0.011551105 = product of:
  0.02310221 = sum of:
    0.02310221 = product of:
      0.04620442 = sum of:
        0.04620442 = weight(_text_:systems in 5243) [ClassicSimilarity], result of:
          0.04620442 = score(doc=5243,freq=4.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.28811008 = fieldWeight in 5243, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.046875 = fieldNorm(doc=5243)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

By "recommending'' Furner refers to collaborative filtering where multiple user rankings of items are used to create a single new ranking for a user, or to a system itself generating rankings of items for its users. This would include document retrieval systems as a subset recommending systems in the second instance, but in the first would make document retrieval system and recommending system synonyms. Information seeking actions are classified either as evaluative (determining the worth of an item), recommending (expressing perceived worth), or informative (examining the content of an item). The task of the information retrieval system is to be to predict the particular ordering that the user would specify in a given context, given complete knowledge of the collection. Citations may be considered as the result of evaluative and recommending decisions by the author, and assigned index terms may be considered as the same sort of decisions by the indexer. The selection of relevant documents by a searcher from a list also involves evaluative and recommending decisions. This suggests that searchers should have the opportunity to bring multiple ranking techniques to bear.

0.011551105 = product of:
  0.02310221 = sum of:
    0.02310221 = product of:
      0.04620442 = sum of:
        0.04620442 = weight(_text_:systems in 703) [ClassicSimilarity], result of:
          0.04620442 = score(doc=703,freq=4.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.28811008 = fieldWeight in 703, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.046875 = fieldNorm(doc=703)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Although user tagging of library resources shows substantial promise as a means of improving the quality of users' access to those resources, several important questions about the level and nature of the warrant for basing retrieval tools on user tagging are yet to receive full consideration by library practitioners and researchers. Among these is the simple evaluative question: What, specifically, are the factors that determine whether or not user-tagging services will be successful? If success is to be defined in terms of the effectiveness with which systems perform the particular functions expected of them (rather than simply in terms of popularity), an understanding is needed both of the multifunctional nature of tagging tools, and of the complex nature of users' mental models of that multifunctionality. In this paper, a conceptual framework is developed for the evaluation of systems that integrate user tagging with more traditional methods of library resource description.

0.010890487 = product of:
  0.021780973 = sum of:
    0.021780973 = product of:
      0.043561947 = sum of:
        0.043561947 = weight(_text_:systems in 395) [ClassicSimilarity], result of:
          0.043561947 = score(doc=395,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.2716328 = fieldWeight in 395, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0625 = fieldNorm(doc=395)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Gives an overview of information retrieval on the WWW. Discusses the characteristics of the digital library and the WWW. Explains information retrieval problems in the context of the WWW, and outlines the responses of developers of information retrieval systems to this problem. Indicates how WWW search services might be improved through the further exploration of ideas developed in the field of library science and artificial intelligence

0.010890487 = product of:
  0.021780973 = sum of:
    0.021780973 = product of:
      0.043561947 = sum of:
        0.043561947 = weight(_text_:systems in 408) [ClassicSimilarity], result of:
          0.043561947 = score(doc=408,freq=8.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.2716328 = fieldWeight in 408, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.03125 = fieldNorm(doc=408)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Content

"Empirical evaluation of knowledge organization (KO) systems, and of the tools and techniques that are used to build systems, is a key component of the system design process: our success in building better systems depends at least partly on our ability to measure the goodness of current systems, and to recognize the factors that affect system performance. The basic evaluative question might be expressed quite simply: How good are the representations or models of the world, of our knowledge of the world, and/ or of expressions of our knowledge of the world - that are produced by our usage of particular KO methods? The straightforwardness of this question is offset by a preliminary need to address metaphysical issues of various kinds, consideration of which can lead us into a quagmire of methodological, epistemological, and ethical problems. What, in this context, is "goodness"? What is the fundamental nature of the kinds of things to be represented? What are the conditions that must be satisfied for a single individual thing to retain its identity over time, and for two individual things to be instances of "the same" kind of thing? Where are the boundaries to be drawn between one thing (or kind of thing) and another? Where does one thing (or kind of thing) stop and another start? How can we come to know the answers to questions about identity, and how we can know when we know? How have we answered questions about identity in different ways at different times and in different places? How ought we to answer questions about identity, and what justifications can we provide in support of our normative claims? As is indicated by the conference organizers' choice of theme for ISKO 2008, designers and evaluators of KO schemes contend on an ongoing basis with issues relating to identity, and a philosophically-informed engagement with such issues is an essential preliminary to understanding evaluation criteria for KO activity. In this talk, the utility for KO of philosophical theories of identity is examined, and motivation is provided for the additional use of such a philosophical framework in evaluating the extent to which KO schemes successfully reflect the cultural identities of their users."

0.010890487 = product of:
  0.021780973 = sum of:
    0.021780973 = product of:
      0.043561947 = sum of:
        0.043561947 = weight(_text_:systems in 3259) [ClassicSimilarity], result of:
          0.043561947 = score(doc=3259,freq=8.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.2716328 = fieldWeight in 3259, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.03125 = fieldNorm(doc=3259)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Empirical evaluation of knowledge organization (KO) systems, and of the tools and techniques that are used to build systems, is a key component of the system design process: our success in building better systems depends at least partly on our ability to measure the goodness of current systems, and to recognize the factors that affect system performance. The basic evaluative question might be expressed quite simply: How good are the representations or models-models of the world, of our knowledge of the world, and/or of expressions of our knowledge of the world-that are produced by our usage of particular KO methods? The straightforwardness of this question is offset by a preliminary need to address metaphysical issues of various kinds, consideration of which can lead us into a quagmire of methodological, epistemological, and ethical problems. What, in this context, is "goodness"? What is the fundamental nature of the kinds of things to be represented? What are the conditions that must be satisfied for a single individual thing to retain its identity over time, and for two individual things to be instances of "the same" kind of thing? Where are the boundaries to be drawn between one thing (or kind of thing) and another? Where does one thing (or kind of thing) stop and another start? How can we come to know the answers to questions about identity, and how we can know when we know? How have we answered questions about identity in different ways at different times and in different places? How ought we to answer questions about identity, and what justifications can we provide in support of our normative claims? As is indicated by the conference organizers' choice of theme for ISKO 2008, designers and evaluators of KO schemes contend on an ongoing basis with issues relating to identity, and a philosophically-informed engagement with such issues is an essential preliminary to understanding evaluation criteria for KO activity. In this talk, the utility for KO of philosophical theories of identity is examined, and motivation is provided for the additional use of such a philosophical framework in evaluating the extent to which KO schemes successfully reflect the cultural identities of their users.

0.010890487 = product of:
  0.021780973 = sum of:
    0.021780973 = product of:
      0.043561947 = sum of:
        0.043561947 = weight(_text_:systems in 3857) [ClassicSimilarity], result of:
          0.043561947 = score(doc=3857,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.2716328 = fieldWeight in 3857, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0625 = fieldNorm(doc=3857)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Folksonomies are indexing languages that emerge from the distributed resource-description activity of multiple agents who make use of online tagging services to assign tags (i.e., category labels) to the resources in collections. Although individuals' motivations for engaging in tagging activity vary widely, folksonomy-based retrieval systems can be evaluated by measuring the degree to which taggers and searchers agree on tag-resource pairings.

0.010890487 = product of:
  0.021780973 = sum of:
    0.021780973 = product of:
      0.043561947 = sum of:
        0.043561947 = weight(_text_:systems in 4705) [ClassicSimilarity], result of:
          0.043561947 = score(doc=4705,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.2716328 = fieldWeight in 4705, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0625 = fieldNorm(doc=4705)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

0.009529176 = product of:
  0.019058352 = sum of:
    0.019058352 = product of:
      0.038116705 = sum of:
        0.038116705 = weight(_text_:systems in 2640) [ClassicSimilarity], result of:
          0.038116705 = score(doc=2640,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.23767869 = fieldWeight in 2640, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0546875 = fieldNorm(doc=2640)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The classification of documents about topics relating to people of mixed race is problematic, partly because of the obscurity of racial categorization in general, and partly because of the limitations and inherent biases of bibliographic classification schemes designed primarily for usage in non-digital environments. Critical race theory is an approach that may prove useful in deterrnining how classification systems such as the Dewey Decimal Classification should most appropriately be stuctured.

0.009529176 = product of:
  0.019058352 = sum of:
    0.019058352 = product of:
      0.038116705 = sum of:
        0.038116705 = weight(_text_:systems in 4714) [ClassicSimilarity], result of:
          0.038116705 = score(doc=4714,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.23767869 = fieldWeight in 4714, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0546875 = fieldNorm(doc=4714)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

For the purposes of this article, the indexing of information is interpreted as the pre-processing of information in order to enable its retrieval. The definition thus spans a dimension extending from classification-based approaches (pre-co-ordinate) to keyword searching (post-co-ordinate). In the first section we clarify our use of terminology, by briefly describing a framework for modelling IR systems in terms of sets of objects, relationships and functions. In the following 3 sections, we discuss the application of indexing functions to document collections of 3 specific types: (1) 'conventional' text databases; (2) hypertext databases; and (3) the World Wide Web, globally distributed across the Internet

0.008837775 = product of:
  0.01767555 = sum of:
    0.01767555 = product of:
      0.0353511 = sum of:
        0.0353511 = weight(_text_:22 in 2754) [ClassicSimilarity], result of:
          0.0353511 = score(doc=2754,freq=2.0), product of:
            0.1827397 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.052184064 = queryNorm
            0.19345059 = fieldWeight in 2754, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2754)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Date

22. 3.2009 18:52:32

0.0068065543 = product of:
  0.013613109 = sum of:
    0.013613109 = product of:
      0.027226217 = sum of:
        0.027226217 = weight(_text_:systems in 6683) [ClassicSimilarity], result of:
          0.027226217 = score(doc=6683,freq=2.0), product of:
            0.16037072 = queryWeight, product of:
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.052184064 = queryNorm
            0.1697705 = fieldWeight in 6683, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.0731742 = idf(docFreq=5561, maxDocs=44218)
              0.0390625 = fieldNorm(doc=6683)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

A textual identity network is a set of documents that share common semantic or linguistic content. For example, the textual identity network of Ben-Hur includes the progenitor work, translations, screen play adaptations, and film performances. A network might also include successively numbered editions, simultaneous editions published in various countries, and other derivative forms. This network expresses how a work evolves over time and through a variety of media. Evolving textual identity can be expressed as a set of relationships among the members of the network. Several taxonomies of intertextual associations have been developed for use in information retrieval systems. The individual documents (books, films, computer files, etc.) contained in a textual identity network can be associated through a number of pairwise relationships, and the network can be studied as a system. This basic pattern makes textual networks ideal candidates for study using network analysis techniques, allowing summary measures that characterize networks. Topological indices provide high-level measures of network structure. This paper concludes on a discussion of how topological indices might be used in document retrieval