Academic
Publications
Solving cluster ensemble problems by bipartite graph partitioning

Solving cluster ensemble problems by bipartite graph partitioning,10.1145/1015330.1015414,Xiaoli Zhang Fern,Carla E. Brodley

Solving cluster ensemble problems by bipartite graph partitioning   (Citations: 82)
BibTex | RIS | RefWorks Download
Cumulative Annual
View Publication
The following links allow you to view full publications. These links are maintained by other sources not affiliated with Microsoft Academic Search.
    • ...1. the feature-based approach that treats the problem of cluster ensembles as the clustering of categorical data (i.e., cluster labels) [5], [6], [40], [47], [48], 2. the direct approach that finds the final partition through relabeling the base clustering results [13], [19], 3. the graph-based approach that employs the graph representation and partitioning technique [7], [12], [44], and 4. the pairwise similarity approach that makes use ...
    • ...Very few attempts (e.g., [12] and [44]) have been made to bring in...
    • ...In addition, the BGP extends the graph-based technique of [12]...
    • ...A number of methods following this approach make use of the graph representation to solve the cluster ensemble problem [7], [12], [44]...
    • ...According to [12], the solution to a cluster ensemble problem is to divide this graph using either METIS or Spectral graph partitioning (SPEC) [39]...
    • ...3.1 Refining BA Matrix through Cluster Relations The proposed approach follows several advanced cluster ensemble methods such as [12] and [44], which apply different consensus functions to the binary cluster-association matrix...
    • ...approximated. Another previous attempt to bring in the similarity between clusters and data points together into consideration has been reported in [12] as the framework of “Hybrid Bipartite Graph Formulation (HBGF).” Despite this useful idea, the final data partition is acquired by applying the spectral graph partitioning algorithm [39] to the bipartite graph that represents the information equivalent to the BA matrix...
    • ...Note that the SPEC graph-partitioning technique has been similarly applied to cluster ensemble problems by Fern and Brodley [12]...
    • ...Graph-based algorithms. The graph-based ensemble methods of [44] (CSPA, HGPA, and MCLA), [12] (HBGF), and [7] (WSPA and WBPA) are employed in this evaluation...
    • ...For comparison, as in [12], [17], and [19], each clustering method divides data points into a partition of K (the number of true classes for each data set) clusters, which is then evaluated against the corresponding true partition using the evaluation indices of: Normalized Mutual Information (NMI) [44], Classification Accuracy (CA) [40], and Rand Index (RI) [42]...
    • ...Brodley for the source code of HBGF [12], and C. Domeniconi for the implementation of LAC [8]...

    Natthakan Iam-Onet al. A Link-Based Approach to the Cluster Ensemble Problem

    • ...Fren and Bordlley [9] developed this algorithm further, designing their hybrid bipartite graph formulation (HBGF)...

    Omar Ayadet al. Heterogeneous ensemble classifier approach for clustering problems

    • ...Real world dataset includes 08X while artificial datasets includes 3-circle, Smile, Half-rings, 2-Spirals ,2 D2k ,8 D5k, EOS, HRCT ,a ndMODIS (Analoui and Sadighian 2006; Azimi et al. 2007; Fern and Brodley 2004; Fred 2001; Fred and Jain 2002; Luo et al. 2007; Ng et al. 2001; Strehl and Ghosh 2002, 2003; Topchy et al. 2003, 2004a,b, 2005)...

    Reza Ghaemiet al. A review: accuracy optimization in clustering ensembles using genetic ...

    • ...Representative algorithms include cluster-based similarity partitioning algorithm (CSPA), hyper-graph partitioning algorithm (HGPA), and metaclustering algorithm (MCLA), all three proposed in [21], as well as hybrid bipartite graph formulation (HBGF) [26]...
    • ...In addition to EAC-AL, we also list the results obtained with three well-known graph-based cluster ensemble algorithms: HBGF [26], CSPA [21], and MCLA [21]...

    Tsaipei Wang. CA-Tree: A Hierarchical Structure for Efficient and Scalable Coassocia...

    • ...Fern and Brodley [31] introduced another cluster ensemble method on graph partitioning named hybrid bipartite graph formulation (HBGF)...

    Jianhua Jiaet al. Soft spectral clustering ensemble applied to image segmentation

Sort by: