Skip to Main content Skip to Navigation
Preprints, Working Papers, ...

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. Paradoxically, this random start is also one of the key weaknesses of these algorithms: nodes are initially connected to dissimilar neighbors, that lie far away according to the similarity metric. As a result, incremental algorithms must first laboriously explore spurious potential neighbors before they can identify similar nodes, and start converging. In this paper, we remove this drawback with Cluster-and-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 random-ness and recursion to pre-cluster 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 speed-ups of up to ×4.42 while incurring only a negligible loss in terms of KNN quality.
Complete list of metadatas

https://hal.inria.fr/hal-02974077
Contributor : Olivier Ruas <>
Submitted on : Wednesday, October 21, 2020 - 3:27:29 PM
Last modification on : Friday, November 27, 2020 - 2:20:11 PM

Files

ClusterAndConquer.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02974077, version 1
  • ARXIV : 2010.11497

Citation

George Giakkoupis, Anne-Marie Kermarrec, Olivier Ruas, François Taïani. Cluster-and-Conquer: When Randomness Meets Graph Locality. 2020. ⟨hal-02974077⟩

Share

Metrics

Record views

26

Files downloads

75