Search

Observed social networks are often considered as proxies for underlying social networks. The analysis of observed networks oftentimes involves the identification of influential nodes via various centrality measures. This paper brings insights from research on adversarial attacks on machine learning systems to the domain of social networks by studying strategies by which an adversary can minimally perturb the observed network structure to achieve their target function of modifying the ranking of a target node according to centrality measures. This can represent the attempt of an adversary to boost or demote the degree to which others perceive individual nodes as influential or powerful. We study the impact of adversarial attacks on targets and victims, and identify metric-based security strategies to mitigate such attacks. We conduct a series of controlled experiments on synthetic network data to identify attacks that allow the adversary to achieve their objective with a single move. We then replicate the experiments with empirical network data. We run our experiments on common network topologies and use common centrality measures. We identify a small set of moves that result in the adversary achieving their objective. This set is smaller for decreasing centrality measures than for increasing them. For both synthetic and empirical networks, we observe that larger networks are less prone to adversarial attacks than smaller ones. Adversarial moves have a higher impact on cellular and small-world networks, while random and scale-free networks are harder to perturb. Also, empirical networks are harder to attack than synthetic networks. Using correlation analysis on our experimental results, we identify how combining measures with low correlation can aid in reducing the effectiveness of adversarial moves. Our results also advance the knowledge about the robustness of centrality measures to network perturbations. The notion of changing social network data to yield adversarial outcomes has practical implications, e.g., for information diffusion on social media, influence and power dynamics in social systems, and developing solutions to improving network security. [ABSTRACT FROM AUTHOR]

Purpose: This study demonstrates that aggregated data from the Repository Analytics and Metrics Portal (RAMP) have significant potential to analyze visibility and use of institutional repositories (IR) as well as potential factors affecting their use, including repository size, platform, content, device and global location. The RAMP dataset is unique and public. Design/methodology/approach: The webometrics methodology was followed to aggregate and analyze use and performance data from 35 institutional repositories in seven countries that were registered with the RAMP for a five-month period in 2019. The RAMP aggregates Google Search Console (GSC) data to show IR items that surfaced in search results from all Google properties. Findings: The analyses demonstrate large performance variances across IR as well as low overall use. The findings also show that device use affects search behavior, that different content types such as electronic thesis and dissertation (ETD) may affect use and that searches originating in the Global South show much higher use of mobile devices than in the Global North. Research limitations/implications: The RAMP relies on GSC as its sole data source, resulting in somewhat conservative overall numbers. However, the data are also expected to be as robot free as can be hoped. Originality/value: This may be the first analysis of aggregate use and performance data derived from a global set of IR, using an openly published dataset. RAMP data offer significant research potential with regard to quantifying and characterizing variances in the discoverability and use of IR content. Peer review: The peer review history for this article is available at: https://publons.com/publon/10.1108/OIR-08-2020-0328 [ABSTRACT FROM AUTHOR]