We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Retrieval

Title: A Comprehensive Survey and Experimental Comparison of Graph-Based Approximate Nearest Neighbor Search

Abstract: Approximate nearest neighbor search (ANNS) constitutes an important operation in a multitude of applications, including recommendation systems, information retrieval, and pattern recognition. In the past decade, graph-based ANNS algorithms have been the leading paradigm in this domain, with dozens of graph-based ANNS algorithms proposed. Such algorithms aim to provide effective, efficient solutions for retrieving the nearest neighbors for a given query. Nevertheless, these efforts focus on developing and optimizing algorithms with different approaches, so there is a real need for a comprehensive survey about the approaches' relative performance, strengths, and pitfalls. Thus here we provide a thorough comparative analysis and experimental evaluation of 13 representative graph-based ANNS algorithms via a new taxonomy and fine-grained pipeline. We compared each algorithm in a uniform test environment on eight real-world datasets and 12 synthetic datasets with varying sizes and characteristics. Our study yields novel discoveries, offerings several useful principles to improve algorithms, thus designing an optimized method that outperforms the state-of-the-art algorithms. This effort also helped us pinpoint algorithms' working portions, along with rule-of-thumb recommendations about promising research directions and suitable algorithms for practitioners in different fields.
Comments: 28 pages, 21 figures, 24 tables, conference
Subjects: Information Retrieval (cs.IR); Databases (cs.DB)
Cite as: arXiv:2101.12631 [cs.IR]
  (or arXiv:2101.12631v2 [cs.IR] for this version)

Submission history

From: Mengzhao Wang [view email]
[v1] Fri, 29 Jan 2021 15:12:35 GMT (390kb,D)
[v2] Mon, 10 May 2021 08:54:16 GMT (2767kb,D)

Link back to: arXiv, form interface, contact.