Search (5 results, page 1 of 1)

0.011883841 = product of:
  0.047535364 = sum of:
    0.047535364 = product of:
      0.09507073 = sum of:
        0.09507073 = weight(_text_:22 in 262) [ClassicSimilarity], result of:
          0.09507073 = score(doc=262,freq=2.0), product of:
            0.17551683 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.050121464 = queryNorm
            0.5416616 = fieldWeight in 262, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.109375 = fieldNorm(doc=262)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

20.10.2000 12:22:23

0.010770457 = product of:
  0.043081827 = sum of:
    0.043081827 = weight(_text_:digital in 4309) [ClassicSimilarity], result of:
      0.043081827 = score(doc=4309,freq=2.0), product of:
        0.19770671 = queryWeight, product of:
          3.944552 = idf(docFreq=2326, maxDocs=44218)
          0.050121464 = queryNorm
        0.21790776 = fieldWeight in 4309, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          3.944552 = idf(docFreq=2326, maxDocs=44218)
          0.0390625 = fieldNorm(doc=4309)
  0.25 = coord(1/4)

Abstract

Semantic annotations have to satisfy quality constraints to be useful for digital libraries, which is particularly challenging on large and diverse datasets. Confidence scores of multi-label classification methods typically refer only to the relevance of particular subjects, disregarding indicators of insufficient content representation at the document-level. Therefore, we propose a novel approach that detects documents rather than concepts where quality criteria are met. Our approach uses a deep, multi-layered regression architecture, which comprises a variety of content-based indicators. We evaluated multiple configurations using text collections from law and economics, where the available content is restricted to very short texts. Notably, we demonstrate that the proposed quality estimation technique can determine subsets of the previously unseen data where considerable gains in document-level recall can be achieved, while upholding precision at the same time. Hence, the approach effectively performs a filtering that ensures high data quality standards in operative information retrieval systems.

0.009946881 = product of:
  0.039787523 = sum of:
    0.039787523 = weight(_text_:library in 4311) [ClassicSimilarity], result of:
      0.039787523 = score(doc=4311,freq=6.0), product of:
        0.1317883 = queryWeight, product of:
          2.6293786 = idf(docFreq=8668, maxDocs=44218)
          0.050121464 = queryNorm
        0.30190483 = fieldWeight in 4311, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          2.6293786 = idf(docFreq=8668, maxDocs=44218)
          0.046875 = fieldNorm(doc=4311)
  0.25 = coord(1/4)

Abstract

Automatic content indexing is one of the innovations that are increasingly changing the way libraries work. In theory, it promises a cataloguing service that would hardly be possible with humans in terms of speed, quantity and maybe quality. The German National Library (DNB) has also recognised this potential and is increasingly relying on the automatic indexing of their catalogue content. The DNB took a major step in this direction in 2017, which was announced in two papers. The announcement was rather restrained, but the content of the papers is all the more explosive for the library community: Since September 2017, the DNB has discontinued the intellectual indexing of series Band H and has switched to an automatic process for these series. The subject indexing of online publications (series O) has been purely automatical since 2010; from September 2017, monographs and periodicals published outside the publishing industry and university publications will no longer be indexed by people. This raises the question: What is the quality of the automatic indexing compared to the manual work or in other words to which degree can the automatic indexing replace people without a signi cant drop in regards to quality?

0.0073997467 = product of:
  0.029598987 = sum of:
    0.029598987 = product of:
      0.059197973 = sum of:
        0.059197973 = weight(_text_:project in 5699) [ClassicSimilarity], result of:
          0.059197973 = score(doc=5699,freq=2.0), product of:
            0.21156175 = queryWeight, product of:
              4.220981 = idf(docFreq=1764, maxDocs=44218)
              0.050121464 = queryNorm
            0.27981415 = fieldWeight in 5699, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.220981 = idf(docFreq=1764, maxDocs=44218)
              0.046875 = fieldNorm(doc=5699)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

The Smart information retrieval project emphazises completely automatic approaches to the understanding and retrieval of large quantities of text. The work in the TREC-2 environment continues, performing both routing and ad hoc experiments. The ad hoc work extends investigations into combining global similarities, giving an overall indication of how a document matches a query, with local similarities identifying a smaller part of the document that matches the query. The performance of ad hoc runs is good, but it is clear that full advantage of the available local information is not been taken advantage of. The routing experiments use conventional relevance feedback approaches to routing, but with a much greater degree of query expansion than was previously done. The length of a query vector is increased by a factor of 5 to 10 by adding terms found in previously seen relevant documents. This approach improves effectiveness by 30-40% over the original query

0.0059419204 = product of:
  0.023767682 = sum of:
    0.023767682 = product of:
      0.047535364 = sum of:
        0.047535364 = weight(_text_:22 in 5001) [ClassicSimilarity], result of:
          0.047535364 = score(doc=5001,freq=2.0), product of:
            0.17551683 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.050121464 = queryNorm
            0.2708308 = fieldWeight in 5001, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0546875 = fieldNorm(doc=5001)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Date

14. 3.1996 13:22:21