A number of indications, such as the number of Nobel Prize winners, show Japan to be a scientifically advanced country. However, standard bibliometric indicators place Japan as a scientifically developing country. The present study is based on the conjecture that scientific publications from Japan belong to two different populations: one originates from studies that advance science and includes highly cited papers, while the other is formed by poorly cited papers with almost zero probability of being highly cited. Although these two categories of papers cannot be easily identified and separated, the scientific level of Japan can be tested by studying the extreme upper tail of the citation distribution of all scientific articles. In contrast to standard bibliometric indicators, which are calculated from the total number of papers or from sets of papers in which the two categories of papers are mixed, in the extreme upper tail, only papers that are addressed to the advance of science will be present. Based on the extreme upper tail, Japan belongs to the group of scientifically advanced countries and is significantly different from countries with a low scientific level. The number of Clarivate Citation laureates also supports our hypothesis that some citation-based metrics do not reveal the high scientific level of Japan. Our findings suggest that Japan is an extreme case of inaccuracy of some citation metrics; the same drawback might affect other countries, although to a lesser degree.
Research collaboration is promoted by governments and research funders but if the relative prevalence and merits of collaboration vary internationally different national and disciplinary strategies may be needed to promote it. This study compares the team size and field normalised citation impact of research across all 27 Scopus broad fields in the ten countries with the most journal articles indexed in Scopus 2008-2012. The results show that team size varies substantially by discipline and country, with Japan (4.2) having two thirds more authors per article than the UK (2.5). Solo authorship …
A multi-parametric family of stretch exponential distributions with various power law tails is introduced and is shown to describe adequately the empirical distributions of scientific citation of individual authors. The four-parametric families are characterized by a normalization coefficient in the exponential part, the power exponent in the power-law asymptotic part, and the coefficient for the transition between the above two parts. The distribution of papers of individual scientist over citations of these papers is studied. Scientists are selected via total number of citations in three ran…
Statistical distributions with heavy tails are ubiquitous in natural and social phenomena. Since the entries in heavy tail have disproportional significance, the knowledge of its exact shape is very important. Citations of scientific papers form one of the best-known heavy tail distributions. Even in this case there is a considerable debate whether citation distribution follows the log-normal or power-law fit. The goal of our study is to solve this debate by measuring citation distribution for a very large and homogeneous data. We measured citation distribution for 418,438 Physics papers publi…
The concept of h-index has been proposed to easily assess a researcher's performance with a single two-dimensional number. However, by using only this single number, we lose significant information about the distribution of the number of citations per article of an author's publication list. Two authors with the same h-index may have totally different distributions of the number of citations per article. One may have a very long "tail" in the citation curve, i.e. he may have published a great number of articles, which did not receive relatively many citations. Another researcher may have a sho…
The research assessments of countries or institutions should reveal their contribution to the advancement of science. Taking into consideration the correlation that exists between scientific impact and number of citations, research assessments can be based on citation counts. However, it is crucial to perform the counts in the heavy tail of citation distributions. Following this heavy tail approach, all measurements performed in Spanish universities show that their research is inefficient because the proportion of highly cited papers is much lower than in the most scientifically advanced count…
Bibliographic records related to tribology research were extracted from SCOPUS and Web of Science databases for the period of 15 years from 1998 to 2012. Macro-level scientometric indicators such as growth rate, share of international collaborative papers, citations per paper, and share of non-cited papers were employed. Further, the Gini coefficient and Simpson Index of Diversity were used. Two new relative indicators : Relative International Collaboration Rate (RICR) and Relative Growth Index (RGI) are proposed in this study. The performance of top countries contributing more than 1000 paper…
The h-index, introduced by Hirsch, is based on the mutual variation between the number of cited and source items. The temporally continuous nature of the citation accretion process causes a shift of cited items from the h-core zone to the adjacent citation-asymmetric zones, viz. h-excess zone, or h-tail zone. The name given to this shifting phenomenon is the Diffusion of Cited Items (DCI). In this paper, two new variables are introduced, i.e., the Fold of Excess citation over Total citations (FET), denoted by $ε^2$ and the Fold of h-core citation over Excess citations (FHE), denoted by $θ^2$. …
This paper presents proof that Google Scholar (GS) can construct documentary sets relevant for evaluating researchers' works. Nobelists in economics were the researchers under analysis, and two types of tests of the GS cites to their works were performed: distributional and semantic. Distributional tests found that the GS cites to the laureates' works conformed to the power-law model with an asymptote or "tail" conterminous with their h-index demarcating their core oeuvre, validating both GS and the h-index. Semantic tests revealed that their works highest in GS cites were on topics for which …
The citations to a set of academic articles are typically unevenly shared, with many articles attracting few citations and few attracting many. It is important to know more precisely how citations are distributed in order to help statistical analyses of citations, especially for sets of articles from a single discipline and a small range of years, as normally used for research evaluation. This article fits discrete versions of the power law, the lognormal distribution and the hooked power law to 20 different Scopus categories, using citations to articles published in 2004 and ignoring uncited …
We analyse the large-scale structure of the journal citation network built from information contained in the Thomson-Reuters Journal Citation Reports. To this end, we take advantage of the network science paraphernalia and explore network properties like density, percolation robustness, average and largest node distances, reciprocity, incoming and outgoing degree distributions, as well as assortative mixing by node degrees. We discover that the journal citation network is a dense, robust, small, and reciprocal world. Furthermore, in and out node degree distributions display long-tails, with fe…
We analyze the citation time-series of manuscripts in three different fields of science; physics, social science and technology. The evolution of the time-series of the yearly number of citations, namely the citation trajectories, diffuse anomalously, their variance scales with time $\propto t^{2H}$, where $H\neq 1/2$. We provide detailed analysis of the various factors that lead to the anomalous behavior: non-stationarity, long-ranged correlations and a fat-tailed increment distribution. The papers exhibit high degree of heterogeneity, across the various fields, as the statistics of the highe…
The h-index provides us with nine natural classes which can be written as a matrix of three vectors. The three vectors are: X=(X1, X2, X3) indicate publication distribution in the h-core, the h-tail, and the uncited ones, respectively; Y=(Y1, Y2, Y3) denote the citation distribution of the h-core, the h-tail and the so-called "excess" citations (above the h-threshold), respectively; and Z=(Z1, Z2, Z3)= (Y1-X1, Y2-X2, Y3-X3). The matrix V=(X,Y,Z)T constructs a measure of academic performance, in which the nine numbers can all be provided with meanings in different dimensions. The "academic trac…
It has recently been claimed by Baccini and coauthors that due to ANVUR's bibliometric evaluations of individuals, departments, and universities, in Italy there has been a surge in self-citations in the last ten years, thus increasing the "inwardness" of Italian research more than has happened abroad. We have studied the database of Ioannidis et al. published on 12 August 2019 of the one hundred thousand most "highly cited" scientists, including about two thousand Italians, and we found that the problem of self-citations in relation to this scientific elite is not significant in Italy, while p…
In-depth Research Impact Summarization through Fine-Grained Temporal Citation Analysis
UKP Lab, TU Darmstadt and Hessian Center for AI (hessian.AI)（图腾实验室，达姆施塔特技术大学及海德堡人工智能中心）；School of Computer Science and Engineering, Hebrew University of Jerusalem（耶路撒冷希伯来大学计算机科学与工程学院）；The Allen Institute for AI (AI2)（人工智能研究所（AI2））
Academic papers have been the protagonists in disseminating expertise. Naturally, paper citation pattern analysis is an efficient and essential way of investigating the knowledge structure of science and technology. For decades, it has been observed that citation of scientific literature follows a heterogeneous and heavy-tailed distribution, and many of them suggest a power-law distribution, log-normal distribution, and related distributions. However, many studies are limited to small-scale approaches; therefore, it is hard to generalize. To overcome this problem, we investigate 21 years of ci…
There are different citation habits in the research fields that influence the obsolescence of the research literature. We analyze the distinctive obsolescence of research literature in disciplinary journals in eight scientific subfields based on cited references distribution, as a synchronous approach. We use both Negative Binomial (NB) and Poisson distributions to capture this obsolescence. The corpus being examined is published in 2019 and covers 22,559 papers citing 872,442 references. Moreover, three measures to analyze the tail of the distribution are proposed: (i) cited reference surviva…
We propose a new citation model which builds on the existing models that explicitly or implicitly include "direct" and "indirect" (learning about a cited paper's existence from references in another paper) citation mechanisms. Our model departs from the usual, unrealistic assumption of uniform probability of direct citation, in which initial differences in citation arise purely randomly. Instead, we demonstrate that a two-mechanism model in which the probability of direct citation is proportional to the number of authors on a paper (team size) is able to reproduce the empirical citation distri…
The citation distribution of a researcher shows the impact of their production and determines the success of their scientific career. However, its application in scientific evaluation is difficult due to the bi-dimensional character of the distribution. Some bibliometric indexes that try to synthesize in a numerical value the principal characteristics of this distribution have been proposed recently. In contrast with other bibliometric measures, the biases that the distribution tails provoke, are reduced by the h-index. However, some limitations in the discrimination among researchers with dif…
As science advances, the academic community has published millions of research papers. Researchers devote time and effort to search relevant manuscripts when writing a paper or simply to keep up with current research. In this paper, we consider the problem of citation recommendation by extending a set of known-to-be-relevant references. Our analysis shows the degrees of cited papers in the subgraph induced by the citations of a paper, called projection graph, follow a power law distribution. Existing popular methods are only good at finding the long tail papers, the ones that are highly connec…
We develop a model for the distribution of scientific citations. The model involves a dual mechanism: in the direct mechanism, the author of a new paper finds an old paper A and cites it. In the indirect mechanism, the author of a new paper finds an old paper A only via the reference list of a newer intermediary paper B, which has previously cited A. By comparison to citation databases, we find that papers having few citations are cited mainly by the direct mechanism. Papers already having many citations ('classics') are cited mainly by the indirect mechanism. The indirect mechanism gives a po…
Whether a scientific paper is cited is related not only to the influence of its author(s) but also to the journal publishing it. Scientists, either proficient or tender, usually submit their most important work to prestigious journals which receives higher citations than the ordinary. How to model the role of scientific journals in citation dynamics is of great importance. In this paper we address this issue through two folds. One is the intrinsic heterogeneity of a paper determined by the impact factor of the journal publishing it. The other is the mechanism of a paper being cited which depen…
Altmetrics promise useful support for assessing the impact of scientific works, including beyond the scholarly community and with very limited citation windows. Unfortunately, altmetrics scores are currently available only for recent articles and cannot be used as covariates in predicting long term impact of publications. However, the study of their statistical properties is a subject of evident interest to scientometricians. Applying the same approaches used in the literature to assess the universality of citation distributions, the intention here is to test whether the universal distribution…
Understanding how a scientist develops new scientific collaborations or how their papers receive new citations is a major challenge in scientometrics. The approach being proposed simultaneously examines the growth processes of the co-authorship and citation networks by analyzing the evolutions of the rich get richer and the fit get richer phenomena. In particular, the preferential attachment function and author fitnesses, which govern the two phenomena, are estimated non-parametrically in each network. The approach is applied to the co-authorship and citation networks of the flagship journal o…
In the double rank analysis of research publications, the local rank position of a country or institution publication is expressed as a function of the world rank position. Excluding some highly or lowly cited publications, the double rank plot fits well with a power law, which can be explained because citations for local and world publications follow lognormal distributions. We report here that the distribution of the number of country or institution publications in world percentiles is a double rank distribution that can be fitted to a power law. Only the data points in high percentiles devi…
Logarithmic transformation of the data has been recommended by the literature in the case of highly skewed distributions such as those commonly found in information science. The purpose of the transformation is to make the data conform to the lognormal law of error for inferential purposes. How does this transformation affect the analysis? We factor analyze and visualize the citation environment of the Journal of the American Chemical Society (JACS) before and after a logarithmic transformation. The transformation strongly reduces the variance necessary for classificatory purposes and therefor…
The ongoing growth in the volume of scientific literature available today precludes researchers from efficiently discerning the relevant from irrelevant content. Researchers are constantly interested in impactful papers, authors and venues in their respective fields. Moreover, they are interested in the so-called recent "rising stars" of these contexts which may lead to attractive directions for future work, collaborations or impactful publication venues. In this work, we address the problem of quantifying research impact in each of these contexts, in order to better direct attention of resear…
We studied the research performance of 69 countries by considering two different types of new knowledge: incremental (normal) and fundamental (radical). In principle, these two types of new knowledge should be assessed at two very different levels of citations, but we demonstrate that a simpler assessment can be performed based on the total number of papers (P) and the ratio of the number of papers in the global top 10% of most cited papers divided to the total number of papers (Ptop 10%/P). P represents the quantity, whereas the Ptop 10%/P ratio represents the efficiency. In ideal countries, …
