Search (59 results, page 2 of 3)

0.0023790773 = product of:
  0.0047581545 = sum of:
    0.0047581545 = product of:
      0.009516309 = sum of:
        0.009516309 = weight(_text_:a in 463) [ClassicSimilarity], result of:
          0.009516309 = score(doc=463,freq=10.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.19940455 = fieldWeight in 463, product of:
              3.1622777 = tf(freq=10.0), with freq of:
                10.0 = termFreq=10.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0546875 = fieldNorm(doc=463)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The author presents a different view on properties of impact measures than given in the paper of De Visscher (2011). He argues that a good impact measure works better when citations are concentrated rather than spread out over articles. The author also presents theoretical evidence that the g-index and the R-index can be close to the square root of the total number of citations, whereas this is not the case for the A-index. Here the author confirms an assertion of De Visscher.

Type

a

0.0022748327 = product of:
  0.0045496654 = sum of:
    0.0045496654 = product of:
      0.009099331 = sum of:
        0.009099331 = weight(_text_:a in 81) [ClassicSimilarity], result of:
          0.009099331 = score(doc=81,freq=28.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.19066721 = fieldWeight in 81, product of:
              5.2915025 = tf(freq=28.0), with freq of:
                28.0 = termFreq=28.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.03125 = fieldNorm(doc=81)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Papers written by several authors can be classified according to the countries of the author affiliations. The empirical part of this paper consists of two datasets. One dataset consists of 1,035 papers retrieved via the search "pedagog*" in the years 2004 and 2005 (up to October) in Academic Search Elite which is a case where phi(m) = the number of papers with m =1, 2,3 ... authors is decreasing, hence most of the papers have a low number of authors. Here we find that #, m = the number of times a country occurs j times in a m-authored paper, j =1, ..., m-1 is decreasing and that # m, m is much higher than all the other #j, m values. The other dataset consists of 3,271 papers retrieved via the search "enzyme" in the year 2005 (up to October) in the same database which is a case of a non-decreasing phi(m): most papers have 3 or 4 authors and we even find many papers with a much higher number of authors. In this case we show again that # m, m is much higher than the other #j, m values but that #j, m is not decreasing anymore in j =1, ..., m-1, although #1, m is (apart from # m, m) the largest number amongst the #j,m. The combinatorial part gives a proof of the fact that #j,m decreases for j = 1, m-1, supposing that all cases are equally possible. This shows that the first dataset is more conform with this model than the second dataset. Explanations for these findings are given. From the data we also find the (we think: new) distribution of number of papers with n =1, 2,3,... countries (i.e. where there are n different countries involved amongst the m (a n) authors of a paper): a fast decreasing function e.g. as a power law with a very large Lotka exponent.

Type

a

0.0022338415 = product of:
  0.004467683 = sum of:
    0.004467683 = product of:
      0.008935366 = sum of:
        0.008935366 = weight(_text_:a in 3993) [ClassicSimilarity], result of:
          0.008935366 = score(doc=3993,freq=12.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.18723148 = fieldWeight in 3993, product of:
              3.4641016 = tf(freq=12.0), with freq of:
                12.0 = termFreq=12.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=3993)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Similarity measures, such as the ones of Jaccard, Dice, or Cosine, measure the similarity between two vectors. A good property for similarity measures would be that, if we add a constant vector to both vectors, then the similarity must increase. We show that Dice and Jaccard satisfy this property while Cosine and both overlap measures do not. Adding a constant vector is called, in Lorenz concentration theory, "nominal increase" and we show that the stronger "transfer principle" is not a required good property for similarity measures. Another good property is that, when we have two vectors and if we add one of these vectors to both vectors, then the similarity must increase. Now Dice, Jaccard, Cosine, and one of the overlap measures satisfy this property, while the other overlap measure does not. Also a variant of this latter property is studied.

Type

a

0.002149515 = product of:
  0.00429903 = sum of:
    0.00429903 = product of:
      0.00859806 = sum of:
        0.00859806 = weight(_text_:a in 2009) [ClassicSimilarity], result of:
          0.00859806 = score(doc=2009,freq=16.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.18016359 = fieldWeight in 2009, product of:
              4.0 = tf(freq=16.0), with freq of:
                16.0 = termFreq=16.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2009)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The article studies the influence of the query formulation of a topic on its h-index. In order to generate pure random sets of documents, we used N-grams (N variable) to measure this influence: strings of zeros, truncated at the end. The used databases are WoS and Scopus. The formula h=T**1/alpha, proved in Egghe and Rousseau (2006) where T is the number of retrieved documents and is Lotka's exponent, is confirmed being a concavely increasing function of T. We also give a formula for the relation between h and N the length of the N-gram: h=D10**(-N/alpha) where D is a constant, a convexly decreasing function, which is found in our experiments. Nonlinear regression on h=T**1/alpha gives an estimation of , which can then be used to estimate the h-index of the entire database (Web of Science [WoS] and Scopus): h=S**1/alpha, , where S is the total number of documents in the database.

Type

a

0.0021279112 = product of:
  0.0042558224 = sum of:
    0.0042558224 = product of:
      0.008511645 = sum of:
        0.008511645 = weight(_text_:a in 6883) [ClassicSimilarity], result of:
          0.008511645 = score(doc=6883,freq=2.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.17835285 = fieldWeight in 6883, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.109375 = fieldNorm(doc=6883)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Type

a

0.0021279112 = product of:
  0.0042558224 = sum of:
    0.0042558224 = product of:
      0.008511645 = sum of:
        0.008511645 = weight(_text_:a in 50) [ClassicSimilarity], result of:
          0.008511645 = score(doc=50,freq=8.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.17835285 = fieldWeight in 50, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0546875 = fieldNorm(doc=50)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Several characteristics of classical Lorenz curves make them unsuitable for the study of a group of topperformers. TOP-curves, defined as a kind of mirror image of TIP-curves used in poverty studies, are shown to possess the properties necessary for adequate empirical ranking of various data arrays, based on the properties of the highest performers (i.e., the core). TOP-curves and essential TOP-curves, also introduced in this article, simultaneously represent the incidence, intensity, and inequality among the top. It is shown that TOPdominance partial order, introduced in this article, is stronger than Lorenz dominance order. In this way, this article contributes to the study of cores, a central issue in applied informetrics.

Type

a

0.0021279112 = product of:
  0.0042558224 = sum of:
    0.0042558224 = product of:
      0.008511645 = sum of:
        0.008511645 = weight(_text_:a in 1881) [ClassicSimilarity], result of:
          0.008511645 = score(doc=1881,freq=8.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.17835285 = fieldWeight in 1881, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0546875 = fieldNorm(doc=1881)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In a previous article, we introduced a general transformation on sources and one on items in an arbitrary information production process (IPP). In this article, we investigate the influence of these transformations on the h-index and on the g-index. General formulae that describe this influence are presented. These are applied to the case that the size-frequency function is Lotkaian (i.e., is a decreasing power function). We further show that the h-index of the transformed IPP belongs to the interval bounded by the two transformations of the h-index of the original IPP, and we also show that this property is not true for the g-index.

Type

a

0.0021060861 = product of:
  0.0042121722 = sum of:
    0.0042121722 = product of:
      0.0084243445 = sum of:
        0.0084243445 = weight(_text_:a in 3267) [ClassicSimilarity], result of:
          0.0084243445 = score(doc=3267,freq=6.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.17652355 = fieldWeight in 3267, product of:
              2.4494898 = tf(freq=6.0), with freq of:
                6.0 = termFreq=6.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0625 = fieldNorm(doc=3267)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Type

a

0.0020392092 = product of:
  0.0040784185 = sum of:
    0.0040784185 = product of:
      0.008156837 = sum of:
        0.008156837 = weight(_text_:a in 1006) [ClassicSimilarity], result of:
          0.008156837 = score(doc=1006,freq=10.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.1709182 = fieldWeight in 1006, product of:
              3.1622777 = tf(freq=10.0), with freq of:
                10.0 = termFreq=10.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=1006)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Fractional frequency distributions of, for example, authors with a certain (fractional) number of papers are very irregular and, therefore, not easy to model or to explain. This article gives a first attempt to this by assuming two simple Lotka laws (with exponent 2): one for the number of authors with n papers (total count here) and one for the number of papers with n authors, n E N. Based an an earlier made convolution model of Egghe, interpreted and reworked now for discrete scores, we are able to produce theoretical fractional frequency distributions with only one parameter, which are in very close agreement with the practical ones as found in a large dataset produced earlier by Rao. The article also shows that (irregular) fractional frequency distributions are a consequence of Lotka's law, and are not examples of breakdowns of this famous historical law.

Type

a

0.0020392092 = product of:
  0.0040784185 = sum of:
    0.0040784185 = product of:
      0.008156837 = sum of:
        0.008156837 = weight(_text_:a in 2004) [ClassicSimilarity], result of:
          0.008156837 = score(doc=2004,freq=10.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.1709182 = fieldWeight in 2004, product of:
              3.1622777 = tf(freq=10.0), with freq of:
                10.0 = termFreq=10.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2004)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

This article studies the h-index (Hirsch index) and the g-index of authors, in case one counts authorship of the cited articles in a fractional way. There are two ways to do this: One counts the citations to these papers in a fractional way or one counts the ranks of the papers in a fractional way as credit for an author. In both cases, we define the fractional h- and g-indexes, and we present inequalities (both upper and lower bounds) between these fractional h- and g-indexes and their corresponding unweighted values (also involving, of course, the coauthorship distribution). Wherever applicable, examples and counterexamples are provided. In a concrete example (the publication citation list of the present author), we make explicit calculations of these fractional h- and g-indexes and show that they are not very different from the unweighted ones.

Type

a

0.0020392092 = product of:
  0.0040784185 = sum of:
    0.0040784185 = product of:
      0.008156837 = sum of:
        0.008156837 = weight(_text_:a in 2366) [ClassicSimilarity], result of:
          0.008156837 = score(doc=2366,freq=10.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.1709182 = fieldWeight in 2366, product of:
              3.1622777 = tf(freq=10.0), with freq of:
                10.0 = termFreq=10.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2366)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Lotka's law was formulated to describe the number of authors with a certain number of publications. Empirical results (Morris & Goldstein, 2007) indicate that Lotka's law is also valid if one counts the number of publications of coauthor pairs. This article gives a simple model proving this to be true, with the same Lotka exponent, if the number of coauthored papers is proportional to the number of papers of the individual coauthors. Under the assumption that this number of coauthored papers is more than proportional to the number of papers of the individual authors (to be explained in the article), we can prove that the size-frequency function of coauthor pairs is Lotkaian with an exponent that is higher than that of the Lotka function of individual authors, a fact that is confirmed in experimental results.

Type

a

0.0020392092 = product of:
  0.0040784185 = sum of:
    0.0040784185 = product of:
      0.008156837 = sum of:
        0.008156837 = weight(_text_:a in 2803) [ClassicSimilarity], result of:
          0.008156837 = score(doc=2803,freq=10.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.1709182 = fieldWeight in 2803, product of:
              3.1622777 = tf(freq=10.0), with freq of:
                10.0 = termFreq=10.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2803)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The relation between Pearson's correlation coefficient and Salton's cosine measure is revealed based on the different possible values of the division of the L1-norm and the L2-norm of a vector. These different values yield a sheaf of increasingly straight lines which together form a cloud of points, being the investigated relation. The theoretical results are tested against the author co-citation relations among 24 informetricians for whom two matrices can be constructed, based on co-citations: the asymmetric occurrence matrix and the symmetric co-citation matrix. Both examples completely confirm the theoretical results. The results enable us to specify an algorithm that provides a threshold value for the cosine above which none of the corresponding Pearson correlations would be negative. Using this threshold value can be expected to optimize the visualization of the vector space.

Type

a

0.0020106873 = product of:
  0.0040213745 = sum of:
    0.0040213745 = product of:
      0.008042749 = sum of:
        0.008042749 = weight(_text_:a in 2067) [ClassicSimilarity], result of:
          0.008042749 = score(doc=2067,freq=14.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.1685276 = fieldWeight in 2067, product of:
              3.7416575 = tf(freq=14.0), with freq of:
                14.0 = termFreq=14.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=2067)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In this paper, for the first time, we present global curves for the measures precision, recall, fallout and miss in function of the number of retrieved documents. Different curves apply for different retrieved systems, for which we give exact definitions in terms of a retrieval density function: perverse retrieval, perfect retrieval, random retrieval, normal retrieval, hereby extending results of Buckland and Gey and of Egghe in the following sense: mathematically more advanced methods yield a better insight into these curves, more types of retrieval are considered and, very importantly, the theory is developed for the "complete" set of measures: precision, recall, fallout and miss. Next we study the interrelationships between precision, recall, fallout and miss in these different types of retrieval, hereby again extending results of Buckland and Gey (incl. a correction) and of Egghe. In the case of normal retrieval we prove that precision in function of recall and recall in function of miss is a concavely decreasing relationship while recall in function of fallout is a concavely increasing relationship. We also show, by producing examples, that the relationships between fallout and precision, miss and precision and miss and fallout are not always convex or concave.

Type

a

0.0018615347 = product of:
  0.0037230693 = sum of:
    0.0037230693 = product of:
      0.0074461387 = sum of:
        0.0074461387 = weight(_text_:a in 3466) [ClassicSimilarity], result of:
          0.0074461387 = score(doc=3466,freq=12.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.15602624 = fieldWeight in 3466, product of:
              3.4641016 = tf(freq=12.0), with freq of:
                12.0 = termFreq=12.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0390625 = fieldNorm(doc=3466)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Power laws as defined in 1926 by A. Lotka are increasing in importance because they have been found valid in varied social networks including the Internet. In this article some unique properties of power laws are proven. They are shown to characterize functions with the scalefree property (also called seif-similarity property) as weIl as functions with the product property. Power laws have other desirable properties that are not shared by exponential laws, as we indicate in this paper. Specifically, Naranan (1970) proves the validity of Lotka's law based on the exponential growth of articles in journals and of the number of journals. His argument is reproduced here and a discrete-time argument is also given, yielding the same law as that of Lotka. This argument makes it possible to interpret the information production process as a seif-similar fractal and show the relation between Lotka's exponent and the (seif-similar) fractal dimension of the system. Lotkaian informetric systems are seif-similar fractals, a fact revealed by Mandelbrot (1977) in relation to nature, but is also true for random texts, which exemplify a very special type of informetric system.

Type

a

0.0018428253 = product of:
  0.0036856506 = sum of:
    0.0036856506 = product of:
      0.0073713013 = sum of:
        0.0073713013 = weight(_text_:a in 3432) [ClassicSimilarity], result of:
          0.0073713013 = score(doc=3432,freq=6.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.1544581 = fieldWeight in 3432, product of:
              2.4494898 = tf(freq=6.0), with freq of:
                6.0 = termFreq=6.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.0546875 = fieldNorm(doc=3432)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

Naranan's important theorem, published in Nature in 1970, states that if the number of journals grows exponentially and if the number of articles in each journal grows exponentially (at the same rate for each journal), then the system satisfies Lotka's law and a formula for the Lotka's exponent is given in function of the growth rates of the journals and the articles. This brief communication re-proves this result by showing that the system satisfies Zipf's law, which is equivalent with Lotka's law. The proof is short and algebraic and does not use infinitesimal arguments.

Type

a

0.001823924 = product of:
  0.003647848 = sum of:
    0.003647848 = product of:
      0.007295696 = sum of:
        0.007295696 = weight(_text_:a in 7119) [ClassicSimilarity], result of:
          0.007295696 = score(doc=7119,freq=2.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.15287387 = fieldWeight in 7119, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.09375 = fieldNorm(doc=7119)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Type

a

0.001823924 = product of:
  0.003647848 = sum of:
    0.003647848 = product of:
      0.007295696 = sum of:
        0.007295696 = weight(_text_:a in 1910) [ClassicSimilarity], result of:
          0.007295696 = score(doc=1910,freq=2.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.15287387 = fieldWeight in 1910, product of:
              1.4142135 = tf(freq=2.0), with freq of:
                2.0 = termFreq=2.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.09375 = fieldNorm(doc=1910)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Type

a

0.001823924 = product of:
  0.003647848 = sum of:
    0.003647848 = product of:
      0.007295696 = sum of:
        0.007295696 = weight(_text_:a in 5154) [ClassicSimilarity], result of:
          0.007295696 = score(doc=5154,freq=8.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.15287387 = fieldWeight in 5154, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=5154)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

The notions aging, obsolescence, impact, growth, utilization, and their relations are studied. It is shown how to correct an observed citation distribution for growth, once the growth distribution is known. The relation of this correction procedure with the calculation of impact measures is explained. More interestingly, we have shown how the influence of growth on aging can be studied over a complete period as a whole. Here, the difference between the so-called average and global aging distributions is the main factor. Our main result is that growth can influence aging but that it does not cause aging. A short overview of some classical articles on this topic is given. Results of these earlier works are placed in the framework set up in this article

Type

a

0.001823924 = product of:
  0.003647848 = sum of:
    0.003647848 = product of:
      0.007295696 = sum of:
        0.007295696 = weight(_text_:a in 3678) [ClassicSimilarity], result of:
          0.007295696 = score(doc=3678,freq=8.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.15287387 = fieldWeight in 3678, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=3678)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In this article concentration (i.e., inequality) aspects of the functions of Zipf and of Lotka are studied. Since both functions are power laws (i.e., they are mathematically the same) it suffices to develop one concentration theory for power laws and apply it twice for the different interpretations of the laws of Zipf and Lotka. After a brief repetition of the functional relationships between Zipf's law and Lotka's law, we prove that Price's law of concentration is equivalent with Zipf's law. A major part of this article is devoted to the development of continuous concentration theory, based an Lorenz curves. The Lorenz curve for power functions is calculated and, based an this, some important concentration measures such as the ones of Gini, Theil, and the variation coefficient. Using Lorenz curves, it is shown that the concentration of a power law increases with its exponent and this result is interpreted in terms of the functions of Zipf and Lotka.

Type

a

0.001823924 = product of:
  0.003647848 = sum of:
    0.003647848 = product of:
      0.007295696 = sum of:
        0.007295696 = weight(_text_:a in 2011) [ClassicSimilarity], result of:
          0.007295696 = score(doc=2011,freq=8.0), product of:
            0.04772363 = queryWeight, product of:
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.041389145 = queryNorm
            0.15287387 = fieldWeight in 2011, product of:
              2.828427 = tf(freq=8.0), with freq of:
                8.0 = termFreq=8.0
              1.153047 = idf(docFreq=37942, maxDocs=44218)
              0.046875 = fieldNorm(doc=2011)
      0.5 = coord(1/2)
  0.5 = coord(1/2)

Abstract

In an earlier paper [Egghe, L. (2004). A universal method of information retrieval evaluation: the "missing" link M and the universal IR surface. Information Processing and Management, 40, 21-30] we showed that, given an IR system, and if P denotes precision, R recall, F fallout and M miss (re-introduced in the paper mentioned above), we have the following relationship between P, R, F and M: P/(1-P)*(1-R)/R*F/(1-F)*(1-M)/M = 1. In this paper we prove the (more difficult) converse: given any four rational numbers in the interval ]0, 1[ satisfying the above equation, then there exists an IR system such that these four numbers (in any order) are the precision, recall, fallout and miss of this IR system. As a consequence we show that any three rational numbers in ]0, 1[ represent any three measures taken from precision, recall, fallout and miss of a certain IR system. We also show that this result is also true for two numbers instead of three.

Type

a