Collaboration Network Analysis Based on Normalized Citation Count and Eigenvector Centrality

Collaboration Network Analysis Based on Normalized Citation Count and Eigenvector Centrality

Anand Bihari (National Institute of Technology Patna, Bihar, India), Sudhakar Tripathi (R. E. C. Ambedkar Nagar, Uttar Pradesh, India) and Akshay Deepak (National Institute of Technology Patna, Bihar, India)
Copyright: © 2019 |Pages: 12
DOI: 10.4018/IJRSDA.2019010104

Abstract

In the research community, the estimation of the scholarly impact of an individual is based on either citation-based indicators or network centrality measures. The network-based centrality measures like degree, closeness, betweenness & eigenvector centrality and the citation-based indicators such as h-index, g-index & i10-index, etc., are used and all of the indicators give full credit to all of the authors of a particular article. This is although the contribution of the authors are different. To determine the actual contribution of an author in a particular article, we have applied arithmetic, geometric and harmonic counting methods for finding the actual contribution of an individual. To find the prominent actor in the network, we have applied eigenvector centrality. To authenticate the proposed analysis, an experimental study has been conducted on 186007 authors collaboration network, that is extracted from IEEE Xplore. The experimental results show that the geometric counting-based credit distribution among scholars gives better results than others.
Article Preview
Top

Newman (2001) discussed the weighted collaboration network of co-authors. Here, authors mentioned that the node represents the individual author and the edge between nodes represents the collaboration and the weight of the edge represents the collaborative strength. Farkas et al. (2007) discussed the weighted collaboration network for appraising the scholarly impact of authors where weight is the geometric mean of citation count earn by the collaborators.

Abbasi et al. (2011) discuss the weighted collaboration of researcher and used social network analysis metrics to evaluate the scientific impact of individuals. In this article, the author used the total number of publications as a collaboration weight between collaborators.

Wang et al. (2011) discuss the weighted co-authorship network and used component analysis, publication frequency and degree centrality for finding the prominent actor in the network.

Liu et al. (2005) formed the collaboration network of digital library research community and proposed a new method for evaluation of the scientific impact of an individual called author rank and mentioned that this method gives a better result than social network analysis metrics. In this article author, consider the sum of the proportional counting of the total number of authors excluding self in a particular paper as a collaboration weight.

Liu et al. (2015) discuss the new method to construct a collaboration network. Here author used geometric series for calculation of share credit to all authors in a particular paper and the collaboration weight computed based on the law of gravity.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 7: 4 Issues (2020): Forthcoming, Available for Pre-Order
Volume 6: 4 Issues (2019): 2 Released, 2 Forthcoming
Volume 5: 4 Issues (2018)
Volume 4: 4 Issues (2017)
Volume 3: 4 Issues (2016)
Volume 2: 2 Issues (2015)
Volume 1: 2 Issues (2014)
View Complete Journal Contents Listing