Ranking nodes in growing networks: When PageRank fails

• DOI: 10.1038/srep16181
• Publication Date: 2015-11-10T10:15:28Z
• Authors: Manuel Sebastian ...; Matúš Medo; Yi-Cheng Zhang
• Keywords: PageRank; Popularity; Rank (graph theory); Ranging

Abstract PageRank is arguably the most popular ranking algorithm which being applied in real systems ranging from information to biological and infrastructure networks. Despite its outstanding popularity broad use different areas of science, relation between algorithm’s efficacy properties network on it acts has not yet been fully understood. We study here PageRank’s performance a model supported by data show that realistic temporal effects make fail individuating valuable nodes for range parameters. Results are qualitative agreement with our model-based findings. This failure reveals static approach filtering inappropriate class growing suggest time-dependent algorithms based linking patterns these needed better rank nodes.