Corpus ID: 1928684

Parallel Correlation Clustering on Big Graphs

  title={Parallel Correlation Clustering on Big Graphs},
  author={Xinghao Pan and Dimitris Papailiopoulos and S. Oymak and B. Recht and K. Ramchandran and Michael I. Jordan},
  • Xinghao Pan, Dimitris Papailiopoulos, +3 authors Michael I. Jordan
  • Published in NIPS 2015
  • Computer Science, Mathematics
  • Given a similarity graph between items, correlation clustering (CC) groups similar items together and dissimilar ones apart. One of the most popular CC algorithms is KwikCluster: an algorithm that serially clusters neighborhoods of vertices, and obtains a 3-approximation ratio. Unfortunately, KwikCluster in practice requires a large number of clustering rounds, a potential bottleneck for large graphs. We present C4 and ClusterWild!, two algorithms for parallel correlation clustering that run… CONTINUE READING
    Motif and Hypergraph Correlation Clustering
    • 9
    • PDF
    Edge partitioning of large graphs
    Fair Correlation Clustering
    • 4
    • PDF
    Parallelism in Randomized Incremental Algorithms
    • 19
    • PDF
    Parallelism in Randomized Incremental Algorithms
    Partial Optimality and Fast Lower Bounds for Weighted Correlation Clustering
    • 5
    • PDF
    Motif clustering and overlapping clustering for social network analysis
    • 27
    • PDF


    Publications referenced by this paper.
    Correlation clustering in general weighted graphs
    • 226
    • PDF
    Correlation clustering with a fixed number of clusters
    • 114
    • PDF
    Correlation clustering in MapReduce
    • 41
    • Highly Influential
    Correlation clustering
    • 113
    • PDF
    Correlation clustering: from theory to practice
    • 18
    • PDF
    Greedy sequential maximal independent set and matching are parallel on average
    • 73
    • PDF
    Correlation Clustering: maximizing agreements via semidefinite programming
    • C. Swamy
    • Computer Science, Mathematics
    • 2004
    • 135
    • PDF
    Clustering with qualitative information
    • 350
    • PDF