Filters








30,678 Hits in 7.4 sec

A new selection strategy for selective cluster ensemble based on Diversity and Independency

Muhammad Yousefnezhad, Ali Reihanian, Daoqiang Zhang, Behrouz Minaei-Bidgoli
2016 Engineering applications of artificial intelligence  
This research introduces a new strategy in cluster ensemble selection by using Independency and Diversity metrics.  ...  In recent years, Diversity and Quality, which are two metrics in evaluation procedure, have been used for selecting basic clustering results in the cluster ensemble selection.  ...  In this paper, a new strategy for evaluating and selecting the best basic results in Cluster Ensemble Selection is introduced. This new strategy is based on the Independency and Diversity metrics. 2.  ... 
doi:10.1016/j.engappai.2016.10.005 fatcat:u6i65qspzrhathe7c55cwhtpyy

Wisdom of Crowds cluster ensemble

Hosein Alizadeh, Muhammad Yousefnezhad, Behrouz Minaei Bidgoli
2015 Intelligent Data Analysis  
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.  ...  Limin et al (2012) used compactness and separation for choosing the reference partition in the cluster ensemble selection. They also used new diversity and quality metrics as a selective strategy.  ... 
doi:10.3233/ida-150728 fatcat:nsughad2fvf27bc5fqx6f5vlru

WoCE: a framework for clustering ensemble by exploiting the wisdom of Crowds theory [article]

Muhammad Yousefnezhad, Sheng-Jun Huang, Daoqiang Zhang
2016 arXiv   pre-print
., diversity, independency, decentralization and aggregation, to guide both the constructing of individual clustering results and the final combination for clustering ensemble.  ...  This paper presents a novel framework for unsupervised and semi-supervised cluster ensemble by exploiting the WOC theory.  ...  ACKNOWLEDGMENT We thank the anonymous reviewers for comments.  ... 
arXiv:1612.06598v1 fatcat:bzn3jt7ldrby5dh6v2ai2677bm

A Review of Cluster Oriented Ensemble Classifier for Improving Performance of Stream Data Classification

Richa Gupta, Hitesh Gupta
2009 INTERNATIONAL JOURNAL OF COMPUTERS & TECHNOLOGY  
Various author proposed a method for stream data classification using ensemble technique.  ...  Cluster oriented ensemble classifier maintain and control the diversity of data during classification process.  ...  For the proper selection of window and horizon used ensemble classifier with support of clustering technique. In ensemble methods, the main strategy is to maintain a dynamic set of classifiers.  ... 
doi:10.24297/ijct.v8i3.3389 fatcat:dmv3x46m5jgjnkatxar4jnwsoi

Less is More: Building Selective Anomaly Ensembles with Application to Event Detection in Temporal Graphs [chapter]

Shebuti Rayana, Leman Akoglu
2015 Proceedings of the 2015 SIAM International Conference on Data Mining  
Ensemble techniques for classification and clustering have long proven effective, yet anomaly ensembles have been barely studied.  ...  Thanks to its selection mechanism, SELECT yields superior performance compared to individual detectors alone, the full ensemble (naively combining all results), and an existing diversity-based ensemble  ...  W911NF-14-1-0029, NSF CAREER 1452425, NSF IIS-1017181, and a gift from Northrop Grumman.  ... 
doi:10.1137/1.9781611974010.70 dblp:conf/sdm/RayanaA15 fatcat:qscscggnbzcq7aqkxf33o7l2be

Taxonomy for characterizing ensemble methods in classification tasks: A review and annotated bibliography

Lior Rokach
2009 Computational Statistics & Data Analysis  
The new taxonomy, presented from the algorithm designer's point of view, is based on five dimensions: inducer, combiner, diversity, size, and members dependency.  ...  This paper presents an updated survey of ensemble methods in classification tasks, while introducing a new taxonomy for characterizing them.  ...  Random-based strategy The most straightforward techniques for creating feature subset-based ensemble are based on random selection or random projection [147] .  ... 
doi:10.1016/j.csda.2009.07.017 fatcat:gemuds4gkffwhg3aschzjh6xpu

The Use of Artificial-Intelligence-Based Ensembles for Intrusion Detection: A Review

Gulshan Kumar, Krishan Kumar
2012 Applied Computational Intelligence and Soft Computing  
However, there is no comprehensive review of ensembles in general and AI-based ensembles for ID to examine and understand their current research status to solve the ID problem.  ...  Artificial-intelligence-(AI-) based techniques play prominent role in development of ensemble for intrusion detection (ID) and have many benefits over other techniques.  ...  Then the method selects a model from each cluster to select subset of available base classifiers. These methods also help to improve diversity of ensemble [17, 45, 88] .  ... 
doi:10.1155/2012/850160 fatcat:rxi5t7appjgl3pn2l7bbb5ru3q

Training Distributed GP Ensemble With a Selective Algorithm Based on Clustering and Pruning for Pattern Classification

G. Folino, C. Pizzuti, G. Spezzano
2008 IEEE Transactions on Evolutionary Computation  
ClustBoostCGP C constructs an ensemble of accurate and diverse classifiers by employing a clustering strategy to each subpopulation located on the nodes of the network.  ...  A boosting algorithm based on cellular genetic programming to build an ensemble of predictors is proposed.  ...  ClustBoostCGP C applies the boosting technique in a distributed hybrid model of parallel GP and uses a clustering-based selective algorithm to maintain the diversity of the ensemble by choosing in each  ... 
doi:10.1109/tevc.2007.906658 fatcat:wrtzkmkbrzdkxf35aamncl36d4

Hybrid Chain-Hypergraph P systems for Multiobjective Ensemble Clustering

Shuo Yan, Yuan Wang, Deting Kong, Jinyan Hu, Jianhua Qu, Xiyu Liu, Jie Xue
2019 IEEE Access  
The global ensemble membrane subsystems conduct a new dense representation multisize ensemble strategy to further improve the accuracy of the final results.  ...  Clustering is a classic combined optimization problem that is widely used in pattern recognition, image processing, market analysis and so on.  ...  We designed three new types of subsystems with new rules and membrane structures to implement multiobjective optimization, increase the population diversity of cluster centers and conduct an ensemble strategy  ... 
doi:10.1109/access.2019.2944675 fatcat:j2l5wxz4yvdjdh5jjhhoi56dle

ESDF: Ensemble Selection using Diversity and Frequency [article]

Shouvick Mondal, Arko Banerjee
2015 arXiv   pre-print
In this paper, we propose an efficient method of ensemble selection for a large ensemble. We prioritize the partitions in the ensemble based on diversity and frequency.  ...  Recently ensemble selection for consensus clustering has emerged as a research problem in Machine Intelligence.  ...  Arun K Pujari of University of Hyderabad and to Prof. Sanghamitra Bandopadhyay of ISI, Kolkata for their valuable suggestions to this research work.  ... 
arXiv:1508.04333v1 fatcat:kk77r6225ncajm73ybhr36uhbq

Classifier Selection by Clustering [chapter]

Hamid Parvin, Behrouz Minaei-Bidgoli, Hamideh Shahpar
2011 Lecture Notes in Computer Science  
Then we partition the classifiers using a clustering algorithm. After that by selecting one classifier per each cluster, we produce the final ensemble.  ...  The proposed method uses bagging and boosting as the generators of base classifiers. Base classifiers are kept fixed as decision trees during the creation of the ensemble.  ...  Since each cluster is produced according to classifiers' outputs, it is highly likely that selecting one classifier from each cluster, and using them as an ensemble can produce a diverse ensemble that  ... 
doi:10.1007/978-3-642-21587-2_7 fatcat:doxkgqgj2vb5nbpomos633k56q

DESIGN OF MULTIPLE CLASSIFIER SYSTEMS [chapter]

Fabio Roli, Giorgio Giacinto
2002 Series in Machine Perception and Artificial Intelligence  
(Woods et al., 1997; Giacinto and Roli, 1997 ) that presented a classifier ensemble approach based on a selection mechanism.  ...  The underlying model could provide some new directions for the development of selection based MCSs. Fig. 1 . 1 1 -A general model for MCS.  ... 
doi:10.1142/9789812778147_0008 fatcat:72nvehqjijbvvfn5j3dfyfrfwu

Less is More: Building Selective Anomaly Ensembles [article]

Shebuti Rayana, Leman Akoglu
2015 arXiv   pre-print
Ensemble techniques for classification and clustering have long proven effective, yet anomaly ensembles have been barely studied.  ...  Thanks to its selection mechanism, SELECT yields superior performance compared to individual detectors alone, the full ensemble (naively combining all results), and an existing diversity-based ensemble  ...  Selecting results based on diversity turns out to be a poor strategy for anomaly ensembles as DivE yields even worse results than the Full ensemble (6/8 in Table 2 ).  ... 
arXiv:1501.01924v1 fatcat:5bkwvf6d7rb4vnfwmetknipmmi

Moderate diversity for better cluster ensembles

Stefan T. Hadjitodorov, Ludmila I. Kuncheva, Ludmila P. Todorova
2006 Information Fusion  
Based on this, a procedure for building a cluster ensemble of a chosen type is proposed (assuming that an ensemble relies on one or more random parameters): generate a small random population of cluster  ...  The results suggest that selection by median diversity is no worse and in some cases is better than building and holding on to one ensemble.  ...  Based on the results, a procedure is suggested for selecting a cluster ensemble from a small population of ensembles.  ... 
doi:10.1016/j.inffus.2005.01.008 fatcat:rp5xi2syqnadjo76wfffrg5hyu

Improving cooperative GP ensemble with clustering and pruning for pattern classification

Gianluigi Folino, Clara Pizzuti, Giandomenico Spezzano
2006 Proceedings of the 8th annual conference on Genetic and evolutionary computation - GECCO '06  
The method evolves a population of trees for a fixed number of rounds and, after each round, it chooses the predictors to include into the ensemble by applying a clustering algorithm to the population  ...  A boosting algorithm based on cellular genetic programming to build an ensemble of predictors is proposed.  ...  ClustBoostCGP C applies the boosting technique in a distributed hybrid multi-island model of parallel GP and uses a clustering-based selective algorithm to maintain the diversity of the ensemble by choosing  ... 
doi:10.1145/1143997.1144139 dblp:conf/gecco/FolinoPS06 fatcat:6xq2ef7iwvbxfme5r4gkk2b4p4
« Previous Showing results 1 — 15 out of 30,678 results