Wisdom of Crowds cluster ensemble

Abstract

Abstract:The Wisdom of Crowds is a phenomenon described in social science that suggests four criteria applicable to groups of people. It is claimed that, if these criteria are satisfied, then the aggregate decisions made by a group will often be better than those of its individual members.Inspired by this concept, we present a novel feedback framework for the cluster ensemble problem, which we call Wisdom of Crowds Cluster Ensemble (WOCCE). Although many conventional cluster ensemble methods focusing on diversity have recently been proposed, WOCCE analyzes the conditions necessary for a crowd to exhibit this collective wisdom. These include decentralization criteria for generating primary results, independence criteria for the base algorithms, and diversity criteria for the ensemble members. We suggest appropriate procedures for evaluating these measures, and propose a new measure to assess the diversity. We evaluate the performance of WOCCE against some other traditional base algorithms as well as state-of-the-art ensemble methods. The results demonstrate the efficiency of WOCCE's aggregate decisionmaking compared to other algorithms.

Dependability‐based cluster weighting in clustering ensemble

“…The clustering ensemble approaches with homogenous clustering algorithms employ a same clustering algorithm during generation of the ensemble pool, that is, all partitions of the ensemble pool are generated by a same clustering algorithm. The partitions of the ensemble pool in homogenous clustering algorithms can be produced by one of the following subtypes: by employing different initializations of a given clustering algorithm , by employing different parameters (like different numbers of clusters) for data clustering using a same clustering algorithm , by employing different data projections for data clustering using a same clustering algorithm , by employing different subsets of dataset features for data clustering using a same clustering algorithm , by employing meta heuristic algorithms for data clustering , and by employing different datasets for data clustering using a same clustering algorithm. …”

“…The whole of empirical investigations are accomplished using Matlab2015. The proposed technique is evaluated against a portion of the best strategies in the field such as: Hybrid Bi‐Partite Graph Formulation ( HB _ PGF ) , Sim‐Rank Similarity ( SRS ) , Weighted‐Connected Triple ( W _ CT ) , Cluster Selection‐Evidence Accumulation Clustering ( CS _ EAC ) , Weighted‐Evidence Accumulation Clustering ( W _ EAC ) , Wisdom of Crowds Ensemble ( WCE ) , Graph Partitioning with Multi‐Granularity Link Analysis ( GPM _ GLA ) , and Two_level Co‐Association Matrix Ensemble ( TCAME ) , Elite Cluster Selection‐Evidence Accumulation Clustering ( ECS _ EAC ) , Cluster‐Level Weighting‐Graph Clustering ( CLW _ GC ) , and Robust Clustering Ensemble based on Iterative Fusion of base Clusters ( RCEIFC ) . These techniques utilize the default suggestions of parameters by their relating authors.…”

WoCE: A framework for Clustering Ensemble by Exploiting the Wisdom of Crowds Theory

“…some of these studies [7], [17], [9], [13] separately run each component of the CES (generate all individual results, then evaluate them, etc.) whereas the rest of studies [18], [12] employed feedback mechanism, which gradually runs each component of the CES (generating the first individual result, then evaluating it, etc.). On the one hand, feedback mechanism uses evaluated the results at each step for improving the quality of the generated results in the next steps.…”

“…For instance, consider a clustering analysis for partitioning emails to normal or spam groups, where the number of instances in the normal group is significantly greater than the number of data points in the spams group. Alizadeh et al [9], [17], [18] proved that the NMI evaluates the similarity between these two clusters equal to 1, while the real similarity is near to zero. This arXiv:1612.06598v1 [stat.ML] 20 Dec 2016 issue can rapidly decrease the performance of the NMI-based CES methods in the big data analysis [17], [9], [18], [12].…”

“…Most of the ensemble methods (especially in the CES) employs the (majority) voting systems [7], [8], [18], [12], such as Boosting and Error-Correcting Output Codes (ECOC) in supervised learning [20] or Evidence Accumulation Clustering (EAC) method in unsupervised learning [19]. Indeed, CES framework just provides a voting system for selecting the robust and stable individual results.…”