Search (2 results, page 1 of 1)

  • × author_ss:"Frei, H.P."
  • × theme_ss:"Retrievalstudien"
  1. Frei, H.P.; Meienberg, S.; Schäuble, P.: ¬The perils of interpreting recall and precision values (1991) 0.00 
    0.004121639 = product of:
  0.016486555 = sum of:
    0.016486555 = weight(_text_:information in 786) [ClassicSimilarity], result of:
      0.016486555 = score(doc=786,freq=6.0), product of:
        0.06134496 = queryWeight, product of:
          1.7554779 = idf(docFreq=20772, maxDocs=44218)
          0.034944877 = queryNorm
        0.2687516 = fieldWeight in 786, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.7554779 = idf(docFreq=20772, maxDocs=44218)
          0.0625 = fieldNorm(doc=786)
  0.25 = coord(1/4)
    Abstract
    The traditional recall and precision measure is inappropriate when retrieval algorithms that retrieve information from Wide Area Networks are evaluated. The principle reason is that information available in WANs is dynamic and its size os orders of magnitude greater than the size of the usual test collections. To overcome these problems, a new efffectiveness measure has been developed, which we call the 'usefulness measure'
    Source
    Information retrieval: GI/GMD-Workshop, Darmstadt, 23.-24.6.1991: Proceedings. Ed.: N. Fuhr
  2. Frei, H.P.; Schäuble, P.: Determining the effectiveness of retrieval algorithms (1991) 0.00 
    0.003091229 = product of:
  0.012364916 = sum of:
    0.012364916 = weight(_text_:information in 787) [ClassicSimilarity], result of:
      0.012364916 = score(doc=787,freq=6.0), product of:
        0.06134496 = queryWeight, product of:
          1.7554779 = idf(docFreq=20772, maxDocs=44218)
          0.034944877 = queryNorm
        0.20156369 = fieldWeight in 787, product of:
          2.4494898 = tf(freq=6.0), with freq of:
            6.0 = termFreq=6.0
          1.7554779 = idf(docFreq=20772, maxDocs=44218)
          0.046875 = fieldNorm(doc=787)
  0.25 = coord(1/4)
    Abstract
    A new effectiveness measure ('usefulness measure') is proposed to circumvent the problems associated with the classical recall and precision measures. It is difficult to evaluate systems that filter extremly dynamic information; the determination of all relevant dodcuments in a real life collection is hardly affordable, and the specification of binary relevance assessments is often problematic. The new measure relies on an statistical approach with which two retrieval algorithms are compared. In contrast to the classical recall and precision measures, the new measure requires only relative judgments, and the reply of the retrieval system os compared directly with the information need of the user rather than with the query. The new measure has the added ability to determine an error probability that indicates haw stable the usefulness measure is. Using a test collection of abstracts from CACM, it is shown that our new measure is also capable of disclosing the effect of manually assigned descriptors and yields a results similar to that of the traditional recall and precision measures.
    Source
    Information processing and management. 27(1991) no.2/3, S.153-164

Authors