Search (3 results, page 1 of 1)

0.0016375937 = product of:
  0.024563905 = sum of:
    0.024563905 = sum of:
      0.0049332716 = weight(_text_:information in 635) [ClassicSimilarity], result of:
        0.0049332716 = score(doc=635,freq=2.0), product of:
          0.050870337 = queryWeight, product of:
            1.7554779 = idf(docFreq=20772, maxDocs=44218)
            0.028978055 = queryNorm
          0.09697737 = fieldWeight in 635, product of:
            1.4142135 = tf(freq=2.0), with freq of:
              2.0 = termFreq=2.0
            1.7554779 = idf(docFreq=20772, maxDocs=44218)
            0.0390625 = fieldNorm(doc=635)
      0.019630633 = weight(_text_:22 in 635) [ClassicSimilarity], result of:
        0.019630633 = score(doc=635,freq=2.0), product of:
          0.101476215 = queryWeight, product of:
            3.5018296 = idf(docFreq=3622, maxDocs=44218)
            0.028978055 = queryNorm
          0.19345059 = fieldWeight in 635, 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=635)
  0.06666667 = coord(1/15)

Date

22. 2.2013 12:03:53

2.848226E-4 = product of:
  0.004272339 = sum of:
    0.004272339 = product of:
      0.008544678 = sum of:
        0.008544678 = weight(_text_:information in 826) [ClassicSimilarity], result of:
          0.008544678 = score(doc=826,freq=6.0), product of:
            0.050870337 = queryWeight, product of:
              1.7554779 = idf(docFreq=20772, maxDocs=44218)
              0.028978055 = queryNorm
            0.16796975 = fieldWeight in 826, product of:
              2.4494898 = tf(freq=6.0), with freq of:
                6.0 = termFreq=6.0
              1.7554779 = idf(docFreq=20772, maxDocs=44218)
              0.0390625 = fieldNorm(doc=826)
      0.5 = coord(1/2)
  0.06666667 = coord(1/15)

Abstract

Several instruments of knowledge organization will reflect different possibilities for information retrieval. In this context, ontologies have a different potential because they allow knowledge discovery, which can be used to retrieve information in a more flexible way. However, this potential can be affected by the theoretical principles adopted in ontology building. The aim of this paper is to discuss, in an introductory way, how a (not exhaustive) set of theoretical principles can influence an aspect of ontologies: their use to obtain inferences. In this context, the role of Ingetraut Dahlberg's Theory of Concept is discussed. The methodology is exploratory, qualitative, and from the technical point of view it uses bibliographic research supported by the content analysis method. It also presents a small example of application as a proof of concept. As results, a discussion about the influence of conceptual definition on subsumption inferences is presented, theoretical contributions are suggested that should be used to guide the formation of hierarchical structures on which such inferences are supported, and examples are provided of how the absence of such contributions can lead to erroneous inferences

2.6310782E-4 = product of:
  0.0039466172 = sum of:
    0.0039466172 = product of:
      0.0078932345 = sum of:
        0.0078932345 = weight(_text_:information in 2754) [ClassicSimilarity], result of:
          0.0078932345 = score(doc=2754,freq=8.0), product of:
            0.050870337 = queryWeight, product of:
              1.7554779 = idf(docFreq=20772, maxDocs=44218)
              0.028978055 = queryNorm
            0.1551638 = fieldWeight in 2754, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.7554779 = idf(docFreq=20772, maxDocs=44218)
              0.03125 = fieldNorm(doc=2754)
      0.5 = coord(1/2)
  0.06666667 = coord(1/15)

Abstract

Belief networks are very appropriate tools to deal with uncertainty. They have been successfully applied to different environments. Information Retrieval has been one of these fields, where this propery (uncertainty) is a very important feature, which has not been well managed by the classical models. In this paper, we present a brief overview of the different applications found in the specialised literature where these graphical models have been used to solve specific problems or to design complete Information Retrieval Models. 1. Introduction The entire Information Retrieval (IR) cycle may be classified as an uncertain process because (Turtle & Croft, 1997): (1) The construction of the representations of documents and queries gives as a result incomplete characterisations, in the form of a set of terms. (2) The submitted query is just a vague description of the users' information need. (3) The computation of the relevance degree of each document with respect to the query inherits the previous uncertainty sources, but also presents the problems that can cause the different representations that a concept may have, as well as, that these concepts are not independent among them. Probabilistic models (Crestani et al., 1998) tried to solve these problems although they have some limitations to overcome them. The development of Belief Networks and their applications to actual problems, with good results, caused that several researchers in the field of IR focused their attention an them. They realised that these kind of networks could be adequate models to be employed in IR, specially designed to work with a high performance in environments in which uncertainty is a very important feature, as the case of IR is. Also, because they can properly represent the relationships among variables.