Search (2 results, page 1 of 1)

0.024313705 = product of:
  0.09725482 = sum of:
    0.09725482 = weight(_text_:evolution in 2853) [ClassicSimilarity], result of:
      0.09725482 = score(doc=2853,freq=4.0), product of:
        0.19585751 = queryWeight, product of:
          5.29663 = idf(docFreq=601, maxDocs=44218)
          0.03697776 = queryNorm
        0.49655905 = fieldWeight in 2853, product of:
          2.0 = tf(freq=4.0), with freq of:
            4.0 = termFreq=4.0
          5.29663 = idf(docFreq=601, maxDocs=44218)
          0.046875 = fieldNorm(doc=2853)
  0.25 = coord(1/4)

Abstract

The objective of this paper is to propose a hierarchical topic evolution model (HTEM) that can organize time-varying topics in a hierarchy and discover their evolutions with multiple timescales. In the proposed HTEM, topics near the root of the hierarchy are more abstract and also evolve in the longer timescales than those near the leaves. To achieve this goal, the distance-dependent Chinese restaurant process (ddCRP) is extended to a new nested process that is able to simultaneously model the dependencies among data and the relationship between clusters. The HTEM is proposed based on the new process for time-stamped documents, in which the timestamp is utilized to measure the dependencies among documents. Moreover, an efficient Gibbs sampler is developed for the proposed HTEM. Our experimental results on two popular real-world data sets verify that the proposed HTEM can capture coherent topics and discover their hierarchical evolutions. It also outperforms the baseline model in terms of likelihood on held-out data.

0.002952157 = product of:
  0.011808628 = sum of:
    0.011808628 = product of:
      0.035425883 = sum of:
        0.035425883 = weight(_text_:22 in 1291) [ClassicSimilarity], result of:
          0.035425883 = score(doc=1291,freq=4.0), product of:
            0.12948982 = queryWeight, product of:
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.03697776 = queryNorm
            0.27358043 = fieldWeight in 1291, product of:
              2.0 = tf(freq=4.0), with freq of:
                4.0 = termFreq=4.0
              3.5018296 = idf(docFreq=3622, maxDocs=44218)
              0.0390625 = fieldNorm(doc=1291)
      0.33333334 = coord(1/3)
  0.25 = coord(1/4)

Abstract

This article studies the impact of differences in citation practices at the subfield, or Web of Science subject category level, using the model introduced in Crespo, Li, and Ruiz-Castillo (2013a), according to which the number of citations received by an article depends on its underlying scientific influence and the field to which it belongs. We use the same Thomson Reuters data set of about 4.4 million articles used in Crespo et al. (2013a) to analyze 22 broad fields. The main results are the following: First, when the classification system goes from 22 fields to 219 subfields the effect on citation inequality of differences in citation practices increases from ?14% at the field level to 18% at the subfield level. Second, we estimate a set of exchange rates (ERs) over a wide [660, 978] citation quantile interval to express the citation counts of articles into the equivalent counts in the all-sciences case. In the fractional case, for example, we find that in 187 of 219 subfields the ERs are reliable in the sense that the coefficient of variation is smaller than or equal to 0.10. Third, in the fractional case the normalization of the raw data using the ERs (or subfield mean citations) as normalization factors reduces the importance of the differences in citation practices from 18% to 3.8% (3.4%) of overall citation inequality. Fourth, the results in the fractional case are essentially replicated when we adopt a multiplicative approach.