Search (4 results, page 1 of 1)

0.013399946 = product of:
  0.053599782 = sum of:
    0.053599782 = weight(_text_:library in 5780) [ClassicSimilarity], result of:
      0.053599782 = score(doc=5780,freq=2.0), product of:
        0.1317883 = queryWeight, product of:
          2.6293786 = idf(docFreq=8668, maxDocs=44218)
          0.050121464 = queryNorm
        0.40671125 = fieldWeight in 5780, product of:
          1.4142135 = tf(freq=2.0), with freq of:
            2.0 = termFreq=2.0
          2.6293786 = idf(docFreq=8668, maxDocs=44218)
          0.109375 = fieldNorm(doc=5780)
  0.25 = coord(1/4)

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.

0.009866329 = product of:
  0.039465316 = sum of:
    0.039465316 = product of:
      0.07893063 = sum of:
        0.07893063 = weight(_text_:project in 3097) [ClassicSimilarity], result of:
          0.07893063 = score(doc=3097,freq=2.0), product of:
            0.21156175 = queryWeight, product of:
              4.220981 = idf(docFreq=1764, maxDocs=44218)
              0.050121464 = queryNorm
            0.37308553 = fieldWeight in 3097, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.220981 = idf(docFreq=1764, maxDocs=44218)
              0.0625 = fieldNorm(doc=3097)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Das Projekt "Aufbau einer Infrastruktur für Information Retrieval-Evaluationen" (AIIRE) bietet eine Softwareinfrastruktur zur Unterstützung von Information Retrieval-Evaluationen (IR-Evaluationen). Die Infrastruktur basiert auf einem Tool-Kit, das bei GESIS im Rahmen des DFG-Projekts IRM entwickelt wurde. Ziel ist es, ein System zu bieten, das zur Forschung und Lehre am Fachbereich Media für IR-Evaluationen genutzt werden kann. This paper describes some aspects of a project called "Aufbau einer Infrastruktur für Information Retrieval-Evaluationen" (AIIRE). Its goal is to build a software-infrastructure which supports the evaluation of information retrieval algorithms.

0.0067679947 = product of:
  0.027071979 = sum of:
    0.027071979 = weight(_text_:library in 3868) [ClassicSimilarity], result of:
      0.027071979 = score(doc=3868,freq=4.0), product of:
        0.1317883 = queryWeight, product of:
          2.6293786 = idf(docFreq=8668, maxDocs=44218)
          0.050121464 = queryNorm
        0.2054202 = fieldWeight in 3868, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          2.6293786 = idf(docFreq=8668, maxDocs=44218)
          0.0390625 = fieldNorm(doc=3868)
  0.25 = coord(1/4)

Abstract

Gross et al. (2015) have demonstrated that about a quarter of hits would typically be lost to keyword searchers if contemporary academic library catalogs dropped their controlled subject headings. This paper reports on an analysis of the loss levels that would result if a bibliographic database, namely the Australian Education Index (AEI), were missing the subject descriptors and identifiers assigned by its professional indexers, employing the methodology developed by Gross and Taylor (2005), and later by Gross et al. (2015). The results indicate that AEI users would lose a similar proportion of hits per query to that experienced by library catalog users: on average, 27% of the resources found by a sample of keyword queries on the AEI database would not have been found without the subject indexing, based on the Australian Thesaurus of Education Descriptors (ATED). The paper also discusses the methodological limitations of these studies, pointing out that real-life users might still find some of the resources missed by a particular query through follow-up searches, while additional resources might also be found through iterative searching on the subject vocabulary. The paper goes on to describe a new research design, based on a before - and - after experiment, which addresses some of these limitations. It is argued that this alternative design will provide a more realistic picture of the value that professionally assigned subject indexing and controlled subject vocabularies can add to literature searching of a more scholarly and thorough kind.