CiteULike: bigbossman's bibliometrics

Stop the numbers game

David Parnas — 2007-10-30T14:36:37-00:00

Commun. ACM, Vol. 50, No. 11. (November 2007), pp. 19-21.

Show me the data

Mike Rossner — 2007-12-18T08:22:14-00:00

J. Cell Biol., Vol. 179, No. 6. (17 December 2007), pp. 1091-1092.

10.1083/jcb.200711140

Clustering of scientific fields by integrating text mining and bibliometrics

Frizo Janssens — 2008-03-11T19:49:39-00:00

(22 May 2007)

Increasing dissemination of scientific and technological publications via the Internet, and their availability in large-scale bibliographic databases, has led to tremendous opportunities to improve classification and bibliometric cartography of science and technology. This metascience benefits from the continuous rise of computing power and the development of new algorithms. Paramount challenges still remain, however. This dissertation verifies the hypothesis that accuracy of clustering and classification of scientific fields is enhanced by incorporation of algorithms and techniques from text mining and bibliometrics. Both textual and bibliometric approaches have advantages and intricacies, and both provide different views on the same interlinked corpus of scientific publications or patents. In addition to textual information in such documents, citations between them also constitute huge networks that yield additional information. We incorporate both points of view and show how to improve on existing text-based and bibliometric methods for the mapping of science. The dissertation is organized into three parts: Firstly, we discuss the use of text mining techniques for information retrieval and for mapping of knowledge embedded in text. We introduce and demonstrate our text mining framework and the use of agglomerative hierarchical clustering. We also investigate the relationship between the number of Latent Semantic Indexing factors, the number of clusters, and clustering performance. Furthermore, we describe a combined semi-automatic strategy to determine the optimal number of clusters in a document set. Secondly, we focus on analysis of large networks that emerge from many individual acts of authors citing other scientific works, or collaborating in the same research endeavor. These networks of science and technology can be analyzed with techniques from bibliometrics and graph theory in order to rank important and relevant entities, for clustering or partitioning, and for extraction of communities. Thirdly, we substantiate the complementarity of text mining and bibliometric methods and we propose schemes for the sound integration of both worlds. The performance of unsupervised clustering and classification significantly improves by deeply merging textual content of scientific publications with the structure of citation graphs. Best results are obtained by a clustering method based on statistical meta-analysis, which significantly outperforms text-based and citation-based solutions. Our hybrid strategies for information retrieval and clustering are corroborated by two case studies. The goal of the first is to unravel and visualize the concept structure of the field of library and information science, and to assess the added value of the hybrid approach. The second study is focused on bibliometric properties, cognitive structure and dynamics of the bioinformatics field. We develop a methodology for dynamic hybrid clustering of evolving bibliographic data sets by matching and tracking clusters through time. To conclude, for the complementary text and graph worlds we devise a hybrid clustering approach that jointly considers both paradigms, and we demonstrate that with an integrated stance we obtain a better interpretation of the structure and evolution of scientific fields.

Combining full text and bibliometric information in mapping scientific disciplines

Patrick Glenisson — 2006-04-12T10:02:26-00:00

Information Processing & Management, Vol. 41, No. 6. (December 2005), pp. 1548-1572.

In the present study results of an earlier pilot study by Glenisson, Glanzel and Persson are extended on the basis of larger sets of papers. Full text analysis and traditional bibliometric methods are serially combined to improve the efficiency of the two individual methods. The text mining methodology already introduced in the pilot study is applied to the complete publication year 2003 of the journal Scientometrics. Altogether 85 documents that can be considered research articles or notes have been selected for this exercise. The outcomes confirm the main results of the pilot study, namely, that such hybrid methodology can be applied to both research evaluation and information retrieval. Nevertheless, Scientometrics documents published in 2003 cover a much broader and more heterogeneous spectrum of bibliometrics and related research than those analysed in the pilot study. A modified subject classification based on the scheme used in an earlier study by Schoepflin and Glanzel has been applied for validation purposes.

Towards mapping library and information science

Frizo Janssens — 2008-03-11T19:42:14-00:00

Information Processing & Management, Vol. 42, No. 6. (December 2006), pp. 1614-1642.

In an earlier study by the authors, full-text analysis and traditional bibliometric methods were combined to map research papers published in the journal Scientometrics. The main objective was to develop appropriate techniques of full-text analysis and to improve the efficiency of the individual methods in the mapping of science. The number of papers was, however, rather limited. In the present study, we extend the quantitative linguistic part of the previous studies to a set of five journals representing the field of Library and Information Science (LIS). Almost 1000 articles and notes published in the period 2002-2004 have been selected for this exercise. The optimum solution for clustering LIS is found for six clusters. The combination of different mapping techniques, applied to the full text of scientific publications, results in a characteristic tripod pattern. Besides two clusters in bibliometrics, one cluster in information retrieval and one containing general issues, webometrics and patent studies are identified as small but emerging clusters within LIS. The study is concluded with the analysis of cluster representations by the selected journals.

Dynamic hybrid clustering of bioinformatics by incorporating text mining and citation analysis

Frizo Janssens — 2008-03-11T19:41:37-00:00

(2007), pp. 360-369.

Comparison of the Hirsch-index with standard bibliometric indicators and with peer judgment for 147 chemistry research groups

Anthony van Raan — 2008-03-10T01:42:34-00:00

Scientometrics, Vol. 67, No. 3. (1 June 2006), pp. 491-502.

In this paper we present characteristics of the statistical correlation between the Hirsch (h-) index and several standard bibliometric indicators, as well as with the results of peer review judgment. We use the results of a large evaluation study of 147 university chemistry research groups in the Netherlands covering the work of about 700 senior researchers during the period 1991-2000. Thus, we deal with research groups rather than individual scientists, as we consider the research group as the most important work floor unit in research, particularly in the natural sciences. Furthermore, we restrict the citation period to a three-year window instead of 'life time counts' in order to focus on the impact of recent work and thus on current research performance. Results show that the h-index and our bibliometric 'crown indicator' both relate in a quite comparable way with peer judgments. But for smaller groups in fields with 'less heavy citation traffic' the crown indicator appears to be a more appropriate measure of research performance.

Co-occurrence matrices and their applications in information science: Extending ACA to the Web environment

Loet Leydesdorff — 2008-03-10T01:42:32-00:00

Journal of the American Society for Information Science and Technology, Vol. 57, No. 12. (2006), pp. 1616-1628.

Co-occurrence matrices, such as cocitation, coword, and colink matrices, have been used widely in the information sciences. However, confusion and controversy have hindered the proper statistical analysis of these data. The underlying problem, in our opinion, involved understanding the nature of various types of matrices. This article discusses the difference between a symmetrical cocitation matrix and an asymmetrical citation matrix as well as the appropriate statistical techniques that can be applied to each of these matrices, respectively. Similarity measures (such as the Pearson correlation coefficient or the cosine) should not be applied to the symmetrical cocitation matrix but can be applied to the asymmetrical citation matrix to derive the proximity matrix. The argument is illustrated with examples. The study then extends the application of co-occurrence matrices to the Web environment, in which the nature of the available data and thus data collection methods are different from those of traditional databases such as the Science Citation Index. A set of data collected with the Google Scholar search engine is analyzed by using both the traditional methods of multivariate analysis and the new visualization software Pajek, which is based on social network analysis and graph theory.

Citation-networks, science landscapes and evolutionary strategies

Andrea Scharnhorst — 2008-03-10T01:42:30-00:00

Scientometrics, Vol. 43, No. 1. (1998), pp. 95-106.

Abstract The construction of virtual science landscapes based on citation networks and the strategic use of the information therein shed new light on the issues of the evolution of the science system and possibilities for control. Citations seem to have a key position in the retrieval and valuation of information from scientific communication networks.Leydesdorff's approach to citation theory takes into account the dual-layered character of communication networks and the second-order nature of the science system. This perspective may help to sharpen the awareness of scientists and science policy makers for possible feedback loops within actions and activities in the science system, and probably nonlinear phenomena resulting therefrom. In this paper an additional link to geometrically oriented evolutionary theories is sketched and a specific landscape concept is used as a framework for some comments.

The matthew index—Concentration patterns and Matthew core journals

M Bonitz — 2008-03-10T01:42:28-00:00

Scientometrics, Vol. 44, No. 3. (1 March 1999), pp. 361-378.

Abstract In this paper we extend our studies to the micro-structure of the Matthew effect for countries (MEC). The MEC allows the ranking of countries by their Matthew Index. The rank distribution of countries, observable only at a macro-level, has its roots in re-distribution processes of citations in every journal of the database. These re-distributed citations we call Matthew citations. Data for 44 countries and 2712 journals (based on theScience Citation Index) are analyzed. The strength of the contribution of the individual journals to the MEC (their number of Matthew citations) is skewly distributed. Due to this high concentration of the MEC we are able to define a new type of journal the Matthew core journal: 145 Matthew core journals account for 50% of the MEC. These journals carry a high potential of gaining a surplus of citations over what is expected and the risk of losing a high number of citations as well.

The Science Strategy Index

M Bonitz — 2008-03-10T01:42:27-00:00

Scientometrics, Vol. 26, No. 1. (26 January 1993), pp. 37-50.

Abstract A new indicator, Science Strategy Index, is proposed, which is based on the scattering of a country's science activity over all science fields and related to the world distribution of the science fields. The indicator allows to compare the structure of the publication output of countries as reflected by the used database, irrespective of the size of the countries.If the science structure of each country is related for comparison to that one of each other country, the indicator converts into a structure measure which enables to cluster countries according to their structural similarity. The cluster map of countries achieved in this way deserves intense discussion upon the different science strategies of countries and their geographic, political, communicative, and socio-cultural background.

Co-citation in the scientific literature: A new measure of the relationship between two documents

Henry Small — 2008-03-10T01:42:22-00:00

Journal of the American Society for Information Science, Vol. 24, No. 4. (1973), pp. 265-269.

A new form of document coupling called co-citation is defined as the frequency with which two documents are cited together. The co-citation frequency of two scientific papers can be determined by comparing lists of citing documents in the Science Citation Index and counting identical entries. Networks of co-cited papers can be generated for specific scientific specialties, and an example is drawn from the literature of particle physics. Co-citation patterns are found to differ significantly from bibliographic coupling patterns, but to agree generally with patterns of direct citation. Clusters of co-cited papers provide a new way to study the specialty structure of science. They may provide a new approach to indexing and to the creation of SDI profiles.

Searching for bridges between disciplines: an author co-citation analysis on the research into scholarly communication

Riitta Karki — 2008-03-10T01:42:24-00:00

Journal of Information Science, Vol. 22, No. 5. (1 January 1996), pp. 323-334.

An author co-citation analysis (ACA) on the research into scholarly communication in sociology of science and in information science within a 20-year period is presented. The question at issue is: to what extent and in what ways the research on scholarly communication brings together the sociology of science and information science, i.e. if the research on scholarly communication acts as a bridge between these two disciplines. It is natural to think of the research on scholarly communication as a common field for these two disciplines, but, by analysing the co-citations accorded to the researchers within both disciplines, one can define the intensity of the relationship or whether it really exists. The ACA suggests that the research of scholarly communication is not enough to be their common denomi nator: sociologists and information scientists mostly stay in their own respective territories. Finally, as the feasibility of ACA is evaluated in the light of the results, the weaknesses of the method become evident. 10.1177/016555159602200501

Cognitive space and information space

Gregory Newby — 2006-08-01T10:09:47-00:00

Journal of the American Society for Information Science and Technology, Vol. 52, No. 12. (2001), pp. 1026-1048.

This article works towards realization of exosomatic memory for information systems. In exosomatic memory systems, the information spaces of systems will be consistent with the cognitive spaces of their human users. A method for measuring concept relations in human cognitive space is presented: the paired comparison survey with Principal Components Analysis. A study to measure the cognitive spaces of 16 research participants is presented. Items measured include relations among seven TREC topic statements as well as 17 concepts from the topic statements. A method for automatically generating information spaces from document collections is presented that uses term cooccurrence, eigensystems analysis, and Principal Components Analysis. The extent of similarity between the cognitive spaces and the information spaces, which were derived independently from each other, is measured. A strong similarity between the information spaces and the cognitive spaces are found, indicating that the methods described may have good utility for working towards information systems that operate as exosomatic memories.

Visualizing a discipline: An author co-citation analysis of information science, 1972-1995

Howard White — 2005-11-18T18:18:42-00:00

Journal of the American Society for Information Science, Vol. 49, No. 4. (7 December 1998), pp. 327-355.

This study presents an extensive domain analysis of a discipline - information science - in terms of its authors. Names of those most frequently cited in 12 key journals from 1972 through 1995 were retrieved from Social Scisearch via DIALOG. The top 120 were submitted to author co-citation analyses, yielding automatic classifications relevant to histories of the field. Tables and graphics reveal: (1) The disciplinary and institutional affiliations of contributors to information science; (2) the specialty structure of the discipline over 24 years; (3) authors' memberships in 1 or more specialties; (4) inertia and change in authors' positions on 2-dimensional subject maps over 3 8-year subperiods, 1972-1979, 1980-1987, 1988-1995; (5) the 2 major subdisciplines of information science and their evolving memberships; (6) <IMG SRC="/giflibrary/12/ldquo.gif" BORDER="0">canonical<IMG SRC="/giflibrary/12/rdquo.gif" BORDER="0"> authors who are in the top 100 in all three subperiods; (7) changes in authors' eminence and influence over the subperiods, as shown by mean co-citation counts; (8) authors with marked changes in their mapped positions over the subperiods; (9) the axes on which authors are mapped, with interpretations; (10) evidence of a paradigm shift in information science in the 1980s; and (11) evidence on the general nature and state of integration of information science. Statistical routines include ALSCAL, INDSCAL, factor analysis, and cluster analysis with SPSS; maps and other graphics were made with DeltaGraph. Theory and methodology are sufficiently detailed to be usable by other researchers. © 1998 John Wiley & Sons, Inc.

Visualizing science by citation mapping

Henry Small — 2005-11-17T20:25:20-00:00

Journal of the American Society for Information Science, Vol. 50, No. 9. (22 June 1999), pp. 799-813.

Science mapping is discussed in the general context of information visualization. Attempts to construct maps of science using citation data are reviewed, focusing on the use of co-citation clusters. New work is reported on a dataset of about 36,000 documents using simplified methods for ordination, and nesting maps hierarchically. An overall map of the dataset shows the multidisciplinary breadth of the document sample, and submaps allow drilling down to the document level. An effort to visualize these data using advanced virtual reality software is described, and the creation of document pathways through the map is seen as a realization of Bush's (1945) associative trails.

Bibliometrics and beyond: some thoughts on web-based citation analysis

Blaise Cronin — 2008-01-25T16:50:37-00:00

Journal of Information Science, Vol. 27, No. 1. (1 February 2001), pp. 1-7.

The idea of a unified citation index to the literature of science was first outlined by Eugene Garfield [1] in 1955 in the journal Science. Science Citation Index has since established itself as the gold standard for scientific information retrieval. It has also become the database of choice for citation analysts and evaluative bibliometricians worldwide. As scientific publication moves to the web, and novel approaches to scholarly communication and peer review establish themselves, new methods of citation and link analysis will emerge to capture often liminal expressions of peer esteem, influence and approbation. The web thus affords bibliometricians rich opportunities to apply and adapt their techniques to new contexts and content: the age of bibliometric spectroscopy' [2] is dawning. 10.1177/016555150102700101

Successive h-indices

András Schubert — 2008-03-05T01:29:41-00:00

Scientometrics, Vol. 70, No. 1. (15 January 2007), pp. 201-205.

Abstract It is suggested that h-indices themselves may form the basis of a series of h-indices at successively higher levels of aggregation. The concept of successive h-indices may usefully contribute to develop a coherent frame for multi-level assessments.

Citation frequency: A biased measure of research impact significantly influenced by the geographical origin of research articles

Gerard Pasterkamp — 2007-11-22T07:02:10-00:00

Scientometrics, Vol. 70, No. 1. (15 January 2007), pp. 153-165.

Abstract Context. The use of citation frequency and impact factor as measures of research quality and journal prestige is being criticized. Citation frequency is augmented by self-citation and for most journals the majority of citations originate from a minority of papers. We hypothesized that citation frequency is also associated with the geographical origin of the research publication. Objective. We determined whether citations originate more frequently from institutes that are located in the same country as the authors of the cited publication than would be expected by chance. Design. We screened citations referring to 1200 cardiovascular publications in the 7 years following their publication. For the 1200 citation recipient publications we documented the country where the research originated (9 countries/regions) and the total number of received citations. For a selection of 8864 citation donor papers we registered the country/region where the citing paper originated. Results. Self-citation was common in cardiovascular journals (n = 1534, 17.8%). After exclusion of self-citation, however, the number of citations that originated from the same country as the author of the citation recipient was found to be on average 31.6% higher than would be expected by chance (p<0.01 for all countries/regions). In absolute numbers, nation oriented citation bias was most pronounced in the USA, the country with the largest research output (p<0.001). Conclusion. Citation frequency was significantly augmented by nation oriented citation bias. This nation oriented citation behaviour seems to mainly influence the cumulative citation number for papers originating from the countries with a larger research output.

Probabilities for encountering genius, basic, ordinary or insignificant papers based on the cumulative nth citation distribution

Leo Egghe — 2008-03-05T01:29:34-00:00

Scientometrics, Vol. 70, No. 1. (15 January 2007), pp. 167-181.

Abstract This article calculates probabilities for the occurrence of different types of papers such as genius papers, basic papers, ordinary papers or insignificant papers. The basis of these calculations are the formulae for the cumulative nth citation distribution, being the cumulative distribution of times at which articles receive their nth(n = 1,2,3,...) citation. These formulae (proved in previous papers) are extended to allow for different aging rates of the papers. These new results are then used to define different importance classes of papers according to the different values of n, in function of time t. Examples are given in case of a classification into four parts: genius papers, basic papers, ordinary papers and (almost) insignificant papers. The fact that, in these examples, the size of each class is inversely related to the importance of the journals in this class is proved in a general mathematical context in which we have an arbitrary number of classes and where the threshold values of n in each class are defined according to the natural law of Weber-Fechner.

Gatekeeper index versus impact factor of science journals

Braun — 2007-06-29T00:41:04-00:00

Scientometrics, Vol. 71, No. 3. (June 2007), pp. 541-543.

Models for citation behavior

Nadarajah — 2007-08-30T16:58:41-00:00

Scientometrics, Vol. 72, No. 2. (August 2007), pp. 291-305.

Generalized Hirsch h-index for disclosing latent facts in citation networks

Sidiropoulos — 2007-08-30T16:58:41-00:00

Scientometrics, Vol. 72, No. 2. (August 2007), pp. 253-280.

Soil science and the index

Budiman Minasny — 2008-03-05T01:29:28-00:00

Scientometrics, Vol. 73, No. 3. (December 2007), pp. 257-264.

Abstract Soil science is a relatively young and specialised field of science. This note discusses the use of the h index as a scientific output measure in soil science. We explore the governing factors of h index in soil science: the number of soil scientists, the number of papers published, the average number of citations, and the age of the scientist. We found the average relationship between h index and scientific age for soil science: h = 0.7 t. The h index for soil science is smaller than other major science disciplines but norms for h need to be established

The declining scientific impact of theses: Implications for electronic thesis and dissertation repositories and graduate studies

Vincent Larivière — 2008-03-05T01:29:29-00:00

Scientometrics, Vol. 74, No. 1. (21 January 2008), pp. 109-121.

Abstract Although the writing of a thesis is a very important step for scientists undertaking a career in research, little information exists on the impact of theses as a source of scientific information. Knowing the impact of theses is relevant not only for students undertaking graduate studies, but also for the building of repositories of electronic theses and dissertations (ETD) and the substantial investment this involves. This paper shows that the impact of theses as information sources has been generally declining over the last century, apart from during the period of the ‘golden years’ of research, 1945 to 1975. There is no evidence of ETDs having a positive impact; on the contrary, since their introduction the impact of theses has actually declined more rapidly. This raises questions about the justification for ETDs and the appropriateness of writing monograph style theses as opposed to publication of a series of peer-reviewed papers as the requirement for fulfilment of graduate studies.

Science and technology in standardization: A statistical analysis of merging knowledge structures

Thilo Gamber — 2008-03-05T01:29:24-00:00

Scientometrics, Vol. 74, No. 1. (21 January 2008), pp. 89-108.

Abstract The objective of this paper is to depict the knowledge array of standards. This is done by identifying and analyzing external effects, specifically spillover effects. The database used is Perinorm. We use a cluster analysis in order to create groups of technology fields for German standards according to the fields of the International Classification of Standards. Methodologically, the distances between these objects or clusters are defined by the chosen distance measure, which in turn is determined by the sum of their cross references. The applied joining clustering method uses these distances between the objects and allows the data to be mapped within a two dimensional space. The results of this mapping show the existence of structures within the standards data fitting to the well-known structure of patent spillovers.

Q-measures for binary divided networks: Bridges between German and English institutes in publications of the Journal of Fluid Mechanics

Lixin Chen — 2008-03-05T01:29:21-00:00

Scientometrics, Vol. 74, No. 1. (21 January 2008), pp. 57-69.

Abstract Q-measures for binary divided networks were introduced in 2004. These measures can value the status of notes as linkage (or bridges) between two groups in a connected undirected network. We collected data from the Web of Science and used a computer programme in order to study Qmeasures for an England-Germany collaboration network in fluid mechanics. The result indicates that Cambridge University, Manchester University, Technische Universität Berlin, the Max Planck Institute, Stuttgart University and Forschungszentrum Karlsruhe play the most important roles as bridges between England and Germany. It is shown that having a high degree centrality and being a key node are important factors explaining the ranking of nodes in a network according to Q-value. It is observed that institutes with a high Q-value have, on average, a higher production than those with a lower Q-value.

Maps of the academic web in the European Higher Education Area — an exploration of visual web indicators

Jose Ortega — 2008-03-05T01:29:08-00:00

Scientometrics, Vol. 74, No. 2. (29 February 2008), pp. 295-308.

Abstract This paper shows maps of the web presence of the European Higher Education Area (EHEA) on the level of universities using hyperlinks and analyses the topology of the European academic network. Its purpose is to combine methods from Social Network Analysis (SNA) and cybermetric techniques in order to ask for tendencies of integration of the European universities visible in their web presence and the role of different universities in the process of the emergence of an European Research Area. We find as a main result that the European network is set up by the aggregation of well-defined national networks, whereby the German and British networks are dominant. The national networks are connected to each other through outstanding national universities in each country.

A new approach to institutional domain analysis: Multilevel research fronts structure

Sandra Miguel — 2008-03-05T01:29:31-00:00

Scientometrics

Abstract The intellectual structure and main research fronts of the Faculty of Natural Sciences and Museum of the National University of La Plata, Argentina is studied, based on the cocitation analysis of subject categories, journals and authors of their scientific publications collected in the Science Citation Index, CD-ROM version, for the period 1991–2000. The objective of this study is to test the utility of those techniques to explore and to visualize the intellectual structure and research fronts of multidisciplinary institutional domains. Special emphasis is laid on the identification of multilevel structures, by means of arrangements of subject categories cocitation analysis and journal cocitation analysis.

Which h-index? — A comparison of WoS, Scopus and Google Scholar

Judit Bar-Ilan — 2008-02-12T21:25:36-00:00

Scientometrics, Vol. 74, No. 2. (29 February 2008), pp. 257-271.

Abstract This paper compares the h-indices of a list of highly-cited Israeli researchers based on citations counts retrieved from the Web of Science, Scopus and Google Scholar respectively. In several case the results obtained through Google Scholar are considerably different from the results based on the Web of Science and Scopus. Data cleansing is discussed extensively.

Locating active actors in the scientific collaboration communities based on interaction topology analyses

Yichuan Jiang — 2008-03-05T01:27:46-00:00

Scientometrics

Abstract While implementing a large-scale research project, it is necessary to appoint some principle scientists, and let each principle scientist lead a research group. In a scientific collaboration community, different scientists perform different roles while they implement the project, and some scientists may be more active than others; these active scientists often undertake the role of leadership or key coordinator in the project. Obviously, we should assign the role of principle scientists onto those active actors in the communities. In this paper, we present the model and algorithms for locating active actors in the community based on the analyses of scientists’ interaction topology, the actors with high connection degrees in the interaction topology can be considered as active ones. Finally, we make some case studies for our model and algorithms.

Bottom-up scientific field detection for dynamical and hierarchical science mapping, methodology and case study

David Chavalarias — 2008-03-05T01:26:37-00:00

Scientometrics

Abstract We propose new methods to detect paradigmatic fields through simple statistics over a scientific content database. We propose an asymmetric paradigmatic proximity metric between terms which provide insight into hierarchical structure of scientific activity and test our methods on a case study with a database made of several millions of resources. We also propose overlapping categorization to describe paradigmatic fields as sets of terms that may have several different usages. Terms can also be dynamically clustered providing a high-level description of the evolution of the paradigmatic fields.

A new methodology for ranking scientific institutions

Jean-Francois Molinari — 2008-03-05T01:26:15-00:00

Scientometrics

Abstract We extend the pioneering work of J. E. Hirsch, the inventor of the h-index, by proposing a simple and seemingly robust approach for comparing the scientific productivity and visibility of institutions. Our main findings are that i) while the h-index is a sensible criterion for comparing scientists within a given field, it does not directly extend to rank institutions of disparate sizes and journals, ii) however, the h-index, which always increases with paper population, has an universal growth rate for large numbers of papers; iii) thus the h-index of a large population of papers can be decomposed into the product of an impact index and a factor depending on the population size, iv) as a complement to the h-index, this new impact index provides an interesting way to compare the scientific production of institutions (universities, laboratories or journals).

Combining concept maps and bibliometric maps: First explorations

RK Buter — 2006-01-18T06:26:15-00:00

Scientometrics, Vol. 66, No. 2. (January 2006), pp. 377-387.

Bibliometric impact measures leveraging topic analysis

Gideon Mann — 2006-09-26T18:30:44-00:00

(2006), pp. 65-74.