Efficient Algorithms for Personalized PageRank Computation: A Survey

Personalized PageRank (PPR) is a traditional measure for node proximity on large graphs. For a pair of nodes <tex-math notation="LaTeX">s</tex-math> and <tex-math notation="LaTeX">t</tex-math> , the PPR value <tex-math notation="LaTeX">πₛ(t)</tex-math> equals the probability that an <tex-math notation="LaTeX">α</tex-math> -discounted random walk from <tex-math notation="LaTeX">s</tex-math> terminates at <tex-math notation="LaTeX">t</tex-math> and reflects the importance between <tex-math notation="LaTeX">s</tex-math> and <tex-math notation="LaTeX">t</tex-math> in a bidirectional way. As a generalization of Google's celebrated PageRank centrality, PPR has been extensively studied and has found multifaceted applications in many fields, such as network analysis, graph mining, and graph machine learning. Despite numerous studies devoted to PPR over the decades, efficient computation of PPR remains a challenging problem, and there is a dearth of systematic summaries and comparisons of existing algorithms. In this paper, we recap several frequently used techniques for PPR computation and conduct a comprehensive survey of various recent PPR algorithms from an algorithmic perspective. We classify these approaches based on the types of queries they address and review their methodologies and contributions. We also discuss some representative algorithms for computing PPR on dynamic graphs and in parallel or distributed environments.