Search (16 results, page 1 of 1)

0.032712005 = product of:
  0.08178001 = sum of:
    0.0538205 = weight(_text_:index in 2934) [ClassicSimilarity], result of:
      0.0538205 = score(doc=2934,freq=2.0), product of:
        0.18579477 = queryWeight, product of:
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.04251826 = queryNorm
        0.28967714 = fieldWeight in 2934, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.046875 = fieldNorm(doc=2934)
    0.027959513 = weight(_text_:system in 2934) [ClassicSimilarity], result of:
      0.027959513 = score(doc=2934,freq=2.0), product of:
        0.13391352 = queryWeight, product of:
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.04251826 = queryNorm
        0.20878783 = fieldWeight in 2934, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.046875 = fieldNorm(doc=2934)
  0.4 = coord(2/5)

Abstract

This comprehensive state-of-the-art book is the first devoted to the important and timely issue of evaluating NLP systems. It addresses the whole area of NLP system evaluation, including aims and scope, problems and methodology. The authors provide a wide-ranging and careful analysis of evaluation concepts, reinforced with extensive illustrations; they relate systems to their environments and develop a framework for proper evaluation. The discussion of principles is completed by a detailed review of practice and strategies in the field, covering both systems for specific tasks, like translation, and core language processors. The methodology lessons drawn from the analysis and review are applied in a series of example cases. A comprehensive bibliography, a subject index, and term glossary are included

0.028704265 = product of:
  0.14352132 = sum of:
    0.14352132 = weight(_text_:index in 5491) [ClassicSimilarity], result of:
      0.14352132 = score(doc=5491,freq=2.0), product of:
        0.18579477 = queryWeight, product of:
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.04251826 = queryNorm
        0.7724724 = fieldWeight in 5491, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.125 = fieldNorm(doc=5491)
  0.2 = coord(1/5)

0.016139356 = product of:
  0.080696784 = sum of:
    0.080696784 = weight(_text_:context in 4422) [ClassicSimilarity], result of:
      0.080696784 = score(doc=4422,freq=2.0), product of:
        0.17622331 = queryWeight, product of:
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.04251826 = queryNorm
        0.45792344 = fieldWeight in 4422, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.078125 = fieldNorm(doc=4422)
  0.2 = coord(1/5)

Abstract

Robertson comments on the theoretical status of IDF term weighting. Its history illustrates how ideas develop in a specific research context, in theory/experiment interaction, and in operational practice.

0.01491174 = product of:
  0.0745587 = sum of:
    0.0745587 = weight(_text_:system in 3156) [ClassicSimilarity], result of:
      0.0745587 = score(doc=3156,freq=2.0), product of:
        0.13391352 = queryWeight, product of:
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.04251826 = queryNorm
        0.5567675 = fieldWeight in 3156, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.125 = fieldNorm(doc=3156)
  0.2 = coord(1/5)

0.014352133 = product of:
  0.07176066 = sum of:
    0.07176066 = weight(_text_:index in 71) [ClassicSimilarity], result of:
      0.07176066 = score(doc=71,freq=2.0), product of:
        0.18579477 = queryWeight, product of:
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.04251826 = queryNorm
        0.3862362 = fieldWeight in 71, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.0625 = fieldNorm(doc=71)
  0.2 = coord(1/5)

Abstract

Examines statistical techniques for exploiting relevance information to weight search terms. These techniques are presented as a natural extension of weighting methods using information about the distribution of index terms in documents in general. A series of relevance weighting functions is derived and is justified by theoretical considerations. In particular, it is shown that specific weighted search methods are implied by a general probabilistic theory of retrieval. Different applications of relevance weighting are illustrated by experimental results for test collections

0.0129114855 = product of:
  0.064557426 = sum of:
    0.064557426 = weight(_text_:context in 2931) [ClassicSimilarity], result of:
      0.064557426 = score(doc=2931,freq=2.0), product of:
        0.17622331 = queryWeight, product of:
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.04251826 = queryNorm
        0.36633876 = fieldWeight in 2931, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.0625 = fieldNorm(doc=2931)
  0.2 = coord(1/5)

Abstract

Automatic summarizing is a research topic whose time has come. The papers illustrate some of the relevant work already under way. Places these papers in their wider context: why research and development on automatic summarizing is timely, what areas of work and ideas it should draw on, how future investigations and experiments can be effectively framed

0.012558117 = product of:
  0.06279058 = sum of:
    0.06279058 = weight(_text_:index in 4420) [ClassicSimilarity], result of:
      0.06279058 = score(doc=4420,freq=2.0), product of:
        0.18579477 = queryWeight, product of:
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.04251826 = queryNorm
        0.33795667 = fieldWeight in 4420, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.0546875 = fieldNorm(doc=4420)
  0.2 = coord(1/5)

Abstract

The exhaustivity of document descriptions and the specificity of index terms are usually regarded as independent. It is suggested that specificity should be interpreted statistically, as a function of term use rather than of term meaning. The effects on retrieval of variations in term specificity are examined, experiments with three test collections showing, in particular, that frequently-occurring terms are required for good overall performance. It is argued that terms should be weighted according to collection frequency, so that matches on less frequent, more specific, terms are of greater value than matches on frequent terms. Results for the test collections show that considerable improvements in performance are obtained with this very simple procedure.

0.01129755 = product of:
  0.05648775 = sum of:
    0.05648775 = weight(_text_:context in 4328) [ClassicSimilarity], result of:
      0.05648775 = score(doc=4328,freq=2.0), product of:
        0.17622331 = queryWeight, product of:
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.04251826 = queryNorm
        0.32054642 = fieldWeight in 4328, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.0546875 = fieldNorm(doc=4328)
  0.2 = coord(1/5)

Abstract

Purpose - This short note seeks to respond to Hjørland and Pederson's paper "A substantive theory of classification for information retrieval" which starts from Sparck Jones's, "Some thoughts on classification for retrieval", originally published in 1970. Design/methodology/approach - The note comments on the context in which the 1970 paper was written, and on Hjørland and Pedersen's views, emphasising the need for well-grounded classification theory and application. Findings - The note maintains that text-based, a posteriori, classification, as increasingly found in applications, is likely to be more useful, in general, than a priori classification. Originality/value - The note elaborates on points made in a well-received earlier paper.

0.008069678 = product of:
  0.040348392 = sum of:
    0.040348392 = weight(_text_:context in 4327) [ClassicSimilarity], result of:
      0.040348392 = score(doc=4327,freq=2.0), product of:
        0.17622331 = queryWeight, product of:
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.04251826 = queryNorm
        0.22896172 = fieldWeight in 4327, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4327)
  0.2 = coord(1/5)

Abstract

The suggestion that classifications for retrieval should be constructed automatically raises some serious problems concerning the sorts of classification which are required, and the way in which formal classification theories should be exploited, given that a retrieval classification is required for a purpose. These difficulties have not been sufficiently considered, and the paper therefore attempts an analysis of them, though no solution of immediate application can be suggested. Starting with the illustrative proposition that a polythetic, multiple, unordered classification is required in automatic thesaurus construction, this is considered in the context of classification in general, where eight sorts of classification can be distinguished, each covering a range of class definitions and class-finding algorithms. The problem which follows is that since there is generally no natural or best classification of a set of objects as such, the evaluation of alternative classifications requires either formal criteria of goodness of fit, or, if a classification is required for a purpose, a precises statement of that purpose. In any case a substantive theory of classification is needed, which does not exist; and since sufficiently precise specifications of retrieval requirements are also lacking, the only currently available approach to automatic classification experiments for information retrieval is to do enough of them

0.008069678 = product of:
  0.040348392 = sum of:
    0.040348392 = weight(_text_:context in 4392) [ClassicSimilarity], result of:
      0.040348392 = score(doc=4392,freq=2.0), product of:
        0.17622331 = queryWeight, product of:
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.04251826 = queryNorm
        0.22896172 = fieldWeight in 4392, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.14465 = idf(docFreq=1904, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4392)
  0.2 = coord(1/5)

Abstract

Purpose - This paper was originally published in 1970 (Journal of documentation. 26(1970), S.89-101), considered the suggestion that classifications for retrieval should be constructed automatically and raised some serious problems concerning the sorts of classification which were required, and the way in which formal classification theories should be exploited, given that a retrieval classification is required for a purpose. These difficulties had not been sufficiently considered, and the paper, therefore, aims to attempt an analysis of them, though no solutions of immediate application could be suggested. Design/methodology/approach - Starting with the illustrative proposition that a polythetic, multiple, unordered classification is required in automatic thesaurus construction, this is considered in the context of classification in general, where eight sorts of classification can be distinguished, each covering a range of class definitions and class-finding algorithms. Findings - Since there is generally no natural or best classification of a set of objects as such, the evaluation of alternative classifications requires either formal criteria of goodness of fit, or, if a classification is required for a purpose, a precise statement of that purpose. In any case a substantive theory of classification is needed, which does not exist; and, since sufficiently precise specifications of retrieval requirements are also lacking, the only currently available approach to automatic classification experiments for information retrieval is to do enough of them. Originality/value - Gives insights into the classification of material for information retrieval.

0.00745587 = product of:
  0.03727935 = sum of:
    0.03727935 = weight(_text_:system in 1916) [ClassicSimilarity], result of:
      0.03727935 = score(doc=1916,freq=2.0), product of:
        0.13391352 = queryWeight, product of:
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.04251826 = queryNorm
        0.27838376 = fieldWeight in 1916, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.0625 = fieldNorm(doc=1916)
  0.2 = coord(1/5)

Abstract

Discusses the TREC programme as a major enterprise in information retrieval research. It reviews its structure as an evaluation exercise, characterises the methods of indexing and retrieval being tested within it in terms of the approaches to system performance factors these represent; analyses the test results for solid, overall conclusions that can be drawn from them; and, in the light of the particular features of the test data, assesses TREC both for generally applicable findings that emerge from it and for directions it offers for future research

0.006523886 = product of:
  0.03261943 = sum of:
    0.03261943 = weight(_text_:system in 4161) [ClassicSimilarity], result of:
      0.03261943 = score(doc=4161,freq=2.0), product of:
        0.13391352 = queryWeight, product of:
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.04251826 = queryNorm
        0.2435858 = fieldWeight in 4161, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.0546875 = fieldNorm(doc=4161)
  0.2 = coord(1/5)

Abstract

Many retrieval experiments have been based on inadequate test collections, and current research is hampered by the lack of proper collections. This short review does not attempt a fully docuemted survey of all the collections used in the past decade: hopefully representative examples have been studied to throw light on the requriements test collections should meet, to show how past collections have been defective, and to suggest guidelines for a future "ideal" test collection. This specifications for this collection can be taken as an indirect comment on our present state of knowledge of major retrieval system variables, and experience in conducting experiments.

0.0062790583 = product of:
  0.03139529 = sum of:
    0.03139529 = weight(_text_:index in 3645) [ClassicSimilarity], result of:
      0.03139529 = score(doc=3645,freq=2.0), product of:
        0.18579477 = queryWeight, product of:
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.04251826 = queryNorm
        0.16897833 = fieldWeight in 3645, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          4.369764 = idf(docFreq=1520, maxDocs=44218)
          0.02734375 = fieldNorm(doc=3645)
  0.2 = coord(1/5)

Abstract

The selection that follows was chosen as it represents "a very early paper an the possibilities allowed by computers an documentation." In the early 1960s computers were being used to provide simple automatic indexing systems wherein keywords were extracted from documents. The problem with such systems was that they lacked vocabulary control, thus documents related in subject matter were not always collocated in retrieval. To improve retrieval by improving recall is the raison d'être of vocabulary control tools such as classifications and thesauri. The question arose whether it was possible by automatic means to construct classes of terms, which when substituted, one for another, could be used to improve retrieval performance? One of the first theoretical approaches to this question was initiated by R. M. Needham and Karen Sparck Jones at the Cambridge Language Research Institute in England.t The question was later pursued using experimental methodologies by Sparck Jones, who, as a Senior Research Associate in the Computer Laboratory at the University of Cambridge, has devoted her life's work to research in information retrieval and automatic naturai language processing. Based an the principles of numerical taxonomy, automatic classification techniques start from the premise that two objects are similar to the degree that they share attributes in common. When these two objects are keywords, their similarity is measured in terms of the number of documents they index in common. Step 1 in automatic classification is to compute mathematically the degree to which two terms are similar. Step 2 is to group together those terms that are "most similar" to each other, forming equivalence classes of intersubstitutable terms. The technique for forming such classes varies and is the factor that characteristically distinguishes different approaches to automatic classification. The technique used by Needham and Sparck Jones, that of clumping, is described in the selection that follows. Questions that must be asked are whether the use of automatically generated classes really does improve retrieval performance and whether there is a true eco nomic advantage in substituting mechanical for manual labor. Several years after her work with clumping, Sparck Jones was to observe that while it was not wholly satisfactory in itself, it was valuable in that it stimulated research into automatic classification. To this it might be added that it was valuable in that it introduced to libraryl information science the methods of numerical taxonomy, thus stimulating us to think again about the fundamental nature and purpose of classification. In this connection it might be useful to review how automatically derived classes differ from those of manually constructed classifications: 1) the manner of their derivation is purely a posteriori, the ultimate operationalization of the principle of literary warrant; 2) the relationship between members forming such classes is essentially statistical; the members of a given class are similar to each other not because they possess the class-defining characteristic but by virtue of sharing a family resemblance; and finally, 3) automatically derived classes are not related meaningfully one to another, that is, they are not ordered in traditional hierarchical and precedence relationships.

0.0055919024 = product of:
  0.027959513 = sum of:
    0.027959513 = weight(_text_:system in 580) [ClassicSimilarity], result of:
      0.027959513 = score(doc=580,freq=2.0), product of:
        0.13391352 = queryWeight, product of:
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.04251826 = queryNorm
        0.20878783 = fieldWeight in 580, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.1495528 = idf(docFreq=5152, maxDocs=44218)
          0.046875 = fieldNorm(doc=580)
  0.2 = coord(1/5)

Abstract

This paper discusses the Text REtrieval Conferences (TREC) programme as a major enterprise in information retrieval research. It reviews its structure as an evaluation exercise, characterises the methods of indexing and retrieval being tested within its terms of the approaches to system performance factors these represent; analyses the test results for solid, overall conclusions that can be drawn from them; and, in the light of the particular features of the test data, assesses TREC both for generally applicable findings that emerge from it and for directions it offers for future research

0.004650343 = product of:
  0.023251716 = sum of:
    0.023251716 = product of:
      0.069755144 = sum of:
        0.069755144 = weight(_text_:29 in 4181) [ClassicSimilarity], result of:
          0.069755144 = score(doc=4181,freq=2.0), product of:
            0.14956595 = queryWeight, product of:
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.04251826 = queryNorm
            0.46638384 = fieldWeight in 4181, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.09375 = fieldNorm(doc=4181)
      0.33333334 = coord(1/3)
  0.2 = coord(1/5)

Date

27.12.2007 19:27:29

0.004650343 = product of:
  0.023251716 = sum of:
    0.023251716 = product of:
      0.069755144 = sum of:
        0.069755144 = weight(_text_:29 in 5092) [ClassicSimilarity], result of:
          0.069755144 = score(doc=5092,freq=2.0), product of:
            0.14956595 = queryWeight, product of:
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.04251826 = queryNorm
            0.46638384 = fieldWeight in 5092, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5176873 = idf(docFreq=3565, maxDocs=44218)
              0.09375 = fieldNorm(doc=5092)
      0.33333334 = coord(1/3)
  0.2 = coord(1/5)

Date

29. 3.1996 18:16:49