Search (2 results, page 1 of 1)

0.008070464 = product of:
  0.032281857 = sum of:
    0.032281857 = product of:
      0.064563714 = sum of:
        0.064563714 = weight(_text_:methods in 4412) [ClassicSimilarity], result of:
          0.064563714 = score(doc=4412,freq=2.0), product of:
            0.18168657 = queryWeight, product of:
              4.0204134 = idf(docFreq=2156, maxDocs=44218)
              0.045191016 = queryNorm
            0.35535768 = fieldWeight in 4412, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.0204134 = idf(docFreq=2156, maxDocs=44218)
              0.0625 = fieldNorm(doc=4412)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Bibliographic searching usually involves finding documents which fulfil already specified search criteria, but Swanson has attempted to develop methods described as a systematic trial and error stzrategy, to find unknown causal scientific connections between already known phenomena, so-called undiscoverd public knowledge, or complementary but disjoint literatures. To this end he conducted 3 investigations in biomedicine. While specialist knowlegde is needed to find the literature linking the literature of problems with that of the solutions, Swanson hopes to make the method more automatic. Clinical trials seem to support his hypothesis

0.0070616566 = product of:
  0.028246626 = sum of:
    0.028246626 = product of:
      0.056493253 = sum of:
        0.056493253 = weight(_text_:methods in 7330) [ClassicSimilarity], result of:
          0.056493253 = score(doc=7330,freq=2.0), product of:
            0.18168657 = queryWeight, product of:
              4.0204134 = idf(docFreq=2156, maxDocs=44218)
              0.045191016 = queryNorm
            0.31093797 = fieldWeight in 7330, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              4.0204134 = idf(docFreq=2156, maxDocs=44218)
              0.0546875 = fieldNorm(doc=7330)
      0.5 = coord(1/2)
  0.25 = coord(1/4)

Abstract

Examines how far automation cluster analysis and association testing are relevant methods for smaller libraries in construction of search thesauri. Using WORDSTAR, a simple form of cluster analysis was tested, with satisfactory results in that the index terms formed clusters of a suitable size. The association test was used to elicit from users' natural language terms which can be used in the search thesaurus as entrance vocabulary. The test showed that users associated very differently in relation to the same stimuli words, with low overlap with terms used in the system's indexing. The results confirmed the need for better feedback. Concludes that while neither method can be used alone, a search thesauri which combine terms from the indexes and from users can be a powerful tool