Cluster-and-Conquer: When Randomness Meets Graph Locality - Inria EPFL Access content directly
Conference Papers Year : 2021

Cluster-and-Conquer: When Randomness Meets Graph Locality

Abstract

K-Nearest-Neighbors (KNN) graphs are central to many emblematic data mining and machine-learning applications. Some of the most efficient KNN graph algorithms are incremental and local: they start from a random graph, which they incrementally improve by traversing neighbors-of-neighbors links. Unfortunately, the initial random graph exhibits a poor graph locality, leading to many unnecessary similarity computations. In this paper, we remove this drawback with Clusterand-Conquer (C 2 for short). Cluster-and-Conquer boosts the starting configuration of greedy algorithms thanks to a novel lightweight clustering mechanism, dubbed FastRandomHash. FastRandomHash leverages randomness and recursion to precluster similar nodes at a very low cost. Our extensive evaluation on real datasets shows that Cluster-and-Conquer significantly outperforms existing approaches, including LSH, yielding speedups of up to ×4.42 and even improving the KNN quality.
Fichier principal
Vignette du fichier
ClusterAndConquer_short_ICDE.pdf (290.81 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03346860 , version 1 (16-09-2021)

Identifiers

Cite

George Giakkoupis, Anne-Marie Kermarrec, Olivier Ruas, François Taïani. Cluster-and-Conquer: When Randomness Meets Graph Locality. ICDE 2021 - IEEE 37th International Conference on Data Engineering, Apr 2021, Chania, Greece. pp.2027-2032, ⟨10.1109/ICDE51399.2021.00195⟩. ⟨hal-03346860⟩
37 View
93 Download

Altmetric

Share

Gmail Facebook X LinkedIn More