As of January 2014, the EBRP is discontinued.

The program has been replaced by the Elsevier metrics development program.


For more information please visit the EMDP website.



 

Granted Proposals: 2013

Team Members

Prof. Wolfgang Glänzel
Katholieke Universiteit Leuven
Wolfgang.Glanzel@econ.kuleuven.be

Project 1. Mathematical Models and Empirical Analysis of Downloads and Citations Processes


Electronic communication and publication have facilitated the access to scholarly literature and accelerated the dissemination of scientific information. The proposed project aims at capturing the observed phenomena by means of bibliometric/informetric analysis and by creating suitable mathematical – statistical models describing those processes. This objective will be achieved using different perspectives and analyzing their interrelation at different levels if aggregation.

Team Members

Israel Mbekezeli Dabengwa
Assistant Librarian, National University of Science & Technology, Zimbabwe
Omwoyo Bosire Onyancha
Associate Professor, University of South Africa, South Africa

Project 2. Predicting Citations to Reproduced Articles in Institutional Repositories: The Zimbabwean Case


The study uses Altmetrics, Bibliometrics and Inferential statistics to predict to what extend research from Zimbabwe has an impact in scholarly publishing. The comparisons use data from Scopus, repositories of five institutes and Altmetrics and measure whether they will be read or cited in Scopus. The aim of this study is also to find reasons why citations to journal articles differ from those in the repositories. It is hoped that the study’s findings will encourage scholars from the developing world to increase their visibility and impact, publish in different types of journals as well as contribute to informal scholarly networks through social media.

Contact information:
Israel Mbekezeli Dabengwa

Team Members

Ivana Roche
INIST-CNRS
Nathalie Antonot
INIST-CNRS
Dominique Besagni
INIST-CNRS
Claire François
INIST-CNRS
Lyne da Sylva
Université de Montréal
Marianne Hörlesberger
AIT, Austrian Institute of Technology GmbH
Edgar Schiebel
AIT, Austrian Institute of Technology GmbH

Project 3. UBIK: Of the importance of identifying the trans-disciplinary vocabulary in the process of Categorization of the concepts related to a scientific domain


In a great number of studies in various fields, there is a growing need for the identification of a trans-disciplinary lexicon. The aim of this study is to develop such lexicon which can considerably decrease an expert’s workload, especially in the stages of evaluation and validation of information extracted from the textual content of records and representing each one in a consistent way. A second pragmatic product of this project is the capacity to associate a given document, represented by its bibliographic record in a database, with a measurement of the level of trans-disciplinarity. Such information would allow the signaling, for instance, of the in-depth specialized documents asking for a deep knowledge on a precise topic or, inversely, documents dealing with subjects undoubtedly related to a particular domain but that could be approached by researchers of the other domains.

Contact information:
Ivana Roche,

Team Members

Jacques Wainer
University of Campinas, Computing Institute

Project 4. Exploration of alternatives for venue classification systems


This study aims to explore alternatives to the link function which assigns a journal to a subject area. While a classification system assign a journal to a subject area by defining a link function, the resulting labeling may not correctly capture journals that are multidisciplinary, and there is no guarantee that even for mono-disciplinary journals, the label will be correct. The investigation into such alternatives includes the use of:

  • Author co-publication (the same author published papers in both venues)
  • Semi-supervised techniques: clustering techniques that assign labels to some of the graph nodes but not all, and let this information aid the clustering process
  • Soft/probabilistic clustering techniques: clustering techniques that assign probabilities to each graph node belonging to a particular label, so one can decide later whether to include or not a node into any cluster

Contact information:
Jacques Wainer

Team Members

Judit Bar-Ilan
Department of Information Science, Bar-Ilan University, Israel

Project 5. Usage, Citation and Readership Data in the Social Sciences


This study aims to explore the relations between usage, citations and readership for a set of Elsevier journals in the Social Sciences and the Humanities. Previous studies showed significant, medium strength correlations between readership counts (derived from the reference manager Mendeley) and citation counts; however the relations with usage data were not studied before. These studies choose to focus on the Humanities as well as specific subareas of the Social Science because they are less studied than the sciences. Social Science publications have reasonable citation counts, and in many fields of Social Sciences, journals are the major publication venue. Therefore, the dataset is expected to be rich and provide insight into these phenomena.

Contact information:
Judit Bar-Ilan

Team Members

Kevin Lanning
Florida Atlantic University
Xingquan Zhu
Florida Atlantic University

Project 6. On the use of first and second-order bibliographic couplings to assess the structure of an institutional network


This study explores the structure of bibliographic couplings with the aim of developing a model of a university populated by persons. This proposed structure may be thought of as a social network in which connections between nodes may reflect shared historical influences, concepts, methods, populations, theoretical approaches, or some combination of these. These links may be considered to be implicitly sourced, as they are based on common references rather than manifest relationships such as co-authorships. This implicit or latent sourcing is an important feature of the model, as it is believed to lead to new synergies between scholars, and new collaborations in both research and instruction.

Contact information:
Kevin Lanning

Team Members

Sylvan Katz
Science and Technology Policy Research, Jubilee Building, University of Sussex

Project 7. Can Cumulative Advantage & Disadvantage Correlations co-exist in a Large Citation Network?


Power law distributions are commonplace in bibliometric research particularly investigations involving citation networks. They indicate an underlying cumulative advantage produced by such things as preferential attachment processes. Power law distributions are a signature of the Matthew Effect in science. It is hypothesized that one reason cumulative disadvantage properties have not been seen may be due to the fact that only total citation counts have been used in current studies. There is evidence suggesting that the inverse Matthew effect maybe found primarily when using between group citations. Also it is hypothesized that by examining how the scaling correlations for the individual components evolve with time it may be possible to deduce how they interact or influence each other to produce the scaling correlation seen for total citation counts. Finally, an independent determination of the within and between group scaling correlations may be used to determine how much each group contributes to its own impact and the impact of other groups.

Contact information:
Sylvan Katz

Team Members

Daniel Torres-Salinas
Center for Applied Medical Research, University of Navarra (Pamplona), Spain
Nicolás Robinson-García
Department of Communication and Information, University of Granada (Granada), Spain
Evaristo Jiménez-Contreras
Department of Communication and Information, University of Granada (Granada), Spain
Rosa Rodríguez-Sánchez (Coordination)
Department of Computer Science and Artificial Intelligence, University de Granada
José A. García
Department of Computer Science and Artificial Intelligence, University de Granada
Joaquín Fdez-Valdivia
Department of Computer Science and Artificial Intelligence, University de Granada

Project 8. Viability of co-downloading data analysis for mapping interdisciplinary research at institutional level


This study focuses on two issue : (1) What do usage (full article downloads)-based indicators measure?’ and (2) The creation and assessment of subject classification systems of scientific-scholarly research that account for multi-disciplinarily. The hypothesis of this study is that an analogy could be established with the relationship between articles tracked by co-citation in the same way that there is a relationship between two closely downloaded articles. By extending the co-citation analysis approach to downloads, this study deepens the concept of co-download analysis as a mean of creating scientific maps similar to those developed through co-citation. The development of such maps would allow a better understanding of the relationships between disciplines as well as compare them with citations.

Contact information:
Daniel Torres-Salinas:
Rosa Rodríguez-Sánchez: