检索结果-内蒙古大学图书馆

clustering ensembles and space discretization - A new regard toward diversity and consensus

PATTERN RECOGNITION LETTERS 2010年第15期31卷 2415-2424页

作者： Montalvao, Jugurta Canuto, Janio Univ Fed Sergipe BR-49100000 Sao Cristovao Brazil Univ Estadual Campinas UNICAMP BR-13083852 Campinas SP Brazil

In recent years, the cluster ensembles have been successfully used to tackle well known drawbacks of individual clustering algorithms. Beyond the expected improvement provided by the averaging effect of many clustering algorithms (clustering committee) aiming at the same goal, some interesting experimental results also show that even committees of completely random partitions may lead to a useful consensus. Another powerful finding in cluster ensemble research is that the blind criterion Averaged Normalized Mutual Information seems to replace actual misclassification ratio, whenever labels are given to actual clusters. In this work, we study what is behind these interesting results and the blind criterion, and we use what we learn from this study to propose a new point of view for analysis and design of clustering committees. The usefulness of this new perspective is illustrated through experimental results. (C) 2010 Elsevier B.V. All rights reserved.

关键词： clustering ensembles Weak partitions ANMI criterion Binary morphology

来源：评论

学校读者我要写书评

暂无评论

clustering ensembles Based on Probability Density Function Estimation 7

Clustering Ensembles Based on Probability Density Function E...

引用

7th IEEE International Conference on Cyber Security and Cloud Computing (CSCloud) / 6th IEEE International Conference on Edge Computing and Scalable Cloud (EdgeCom)

作者： Wu, Yingyan He, Yulin Huang, Joshua Zhexue Shenzhen Univ Coll Comp Sci & Software Engn Shenzhen 518060 Peoples R China

ISBN: (纸本)9781728165509

The traditional clustering algorithms rely excessively on the similarity of the geographic distance between objects, like DBSCAN, which is unlikely to handle uncertain objects that are geometrically indistinguishable. Attempting to prevent this reliance, a model that adopts a probability density estimation function for clustering ensembles is therefore proposed. In other words, the object would be assigned to its most appropriate cluster through mining the distribution of data in the base clusterings fully considering the relationship between the objects, and finding the probability density function that fits datasets. It is worth mentioning that the approach of generating base clusterings through data sampling not only relatively simple to extract potential relationships between objects due to random combinations but also is beneficial to cluster large data sets in the real world. It is first to extract a certain proportion of samples from the dataset, and subsequently to run the k-means algorithm to generate base clusterings repeatedly. Accordingly, an object-cluster association matrix set was conducted, which has summarized the base clusterings, namely the binary association matrix. Under the inspiration of the Bayesian classifier, the object is divided into its corresponding clusters determined by the prior probabilities and posterior probabilities of clusters. Secondly, we can easily obtain the discrete probability distributions of the data objects under the condition of each object's label from the matrix, which is posterior probabilities of clusters, so one important point left in this paper is to estimate the probability distributions of object labels. To address this, a well-known kernel density estimation method was adopted to attain the cluster probability distributions. Finally, the data objects are assigned to their most likely clusters via calculating the Bayesian formula. Extensive experimental results of the proposed method on several datasets were c

关键词： clustering ensembles base clustering probability density function probability distribution kernel density estimation

来源：评论

学校读者我要写书评

暂无评论

Steganalysis Over Large-Scale Social Networks With High-Order Joint Features and clustering ensembles

引用

IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY 2016年第2期11卷 344-357页

作者： Li, Fengyong Wu, Kui Lei, Jingsheng Wen, Mi Bi, Zhongqin Gu, Chunhua Shanghai Univ Elect Power Coll Comp Sci & Technol Shanghai 200090 Peoples R China Univ Victoria Dept Comp Sci Victoria BC V8W 3P6 Canada

This paper tackles a recent challenge in identifying culprit actors, who try to hide confidential payload with steganography, among many innocent actors in social media networks. The problem is called steganographer detection problem and is significantly different from the traditional stego detection problem that classifies an individual object as a cover or a stego. To solve the steganographer detection problem over large-scale social media networks, this paper proposes a method that uses high-order joint features and clustering ensembles. It employs 250-D features calculated from the high-order joint matrices of Discrete Cosine Transform (DCT) coefficients of JPEG images, which indicate the dependencies of image content. Furthermore, a number of hierarchical sub-clusterings trained by the features are integrated as a clustering ensemble based on the majority voting strategy, which is used to make optimal decisions on suspicious steganographers. Experimental results show that the proposed scheme is effective and efficient in identifying potential steganographers in large-scale social media networks, and has better performance when tested against the state-of-the-art steganographic methods.

关键词： Steganalysis social networks JPEG image high-order joint features clustering ensembles

来源：评论

学校读者我要写书评

暂无评论

Hybrid sampling on mutual information entropy-based clustering ensembles for optimizations

引用

NEUROCOMPUTING 2010年第7-9期73卷 1457-1464页

作者： Wang, Feng Yang, Cheng Lin, Zhiyi Li, Yuanxiang Yuan, Yuan Wuhan Univ State Key Lab Software Engn Wuhan 430072 Peoples R China Guangdong Univ Technol Fac Comp Guangzhou Guangdong Peoples R China Aston Univ Sch Engn & Appl Sci Birmingham B4 7ET W Midlands England

In this paper, we focus on the design of bivariate EDAs for discrete optimization problems and propose a new approach named HSMIEC. While the current EDAs require much time in the statistical learning process as the relationships among the variables are too complicated, we employ the Selfish gene theory (SG) in this approach, as well as a Mutual Information and Entropy based Cluster (MIEC) model is also set to optimize the probability distribution of the virtual population. This model uses a hybrid sampling method by considering both the clustering accuracy and clustering diversity and an incremental learning and resample scheme is also set to optimize the parameters of the correlations of the variables. Compared with several benchmark problems, our experimental results demonstrate that HSMIEC often performs better than some other EDAs, such as BMDA, COMIT, MIMIC and ECGA. (C) 2009 Elsevier B.V. All rights reserved.

关键词： Estimation of distribution algorithm Selfish gene theory Mutual information entropy clustering ensembles

来源：评论

学校读者我要写书评

暂无评论

Multi-objective design of hierarchical consensus functions for clustering ensembles via genetic programming

引用

DECISION SUPPORT SYSTEMS 2011年第4期51卷 794-809页

作者： Coelho, Andre L. V. Fernandes, Everlandio Faceli, Katti Univ Fortaleza Ctr Technol Sci Grad Program Appl Informat BR-60811905 Fortaleza CE Brazil Univ Fed Sao Carlos BR-18052780 Sorocaba SP Brazil

This paper investigates a genetic programming (GP) approach aimed at the multi-objective design of hierarchical consensus functions for clustering ensembles. By this means, data partitions obtained via different clustering techniques can be continuously refined (via selection and merging) by a population of fusion hierarchies having complementary validation indices as objective functions. To assess the potential of the novel framework in terms of efficiency and effectiveness, a series of systematic experiments, involving eleven variants of the proposed GP-based algorithm and a comparison with basic as well as advanced clustering methods (of which some are clustering ensembles and/or multi-objective in nature), have been conducted on a number of artificial, benchmark and bioinformatics datasets. Overall, the results corroborate the perspective that having fusion hierarchies operating on well-chosen subsets of data partitions is a fine strategy that may yield significant gains in terms of clustering robustness. (C) 2011 Elsevier B.V. All rights reserved.

关键词： Cluster analysis clustering ensembles Multi-objective clustering Hierarchical fusion Partition selection Genetic programming

来源：评论

学校读者我要写书评

暂无评论

Projective clustering ensembles

引用

DATA MINING AND KNOWLEDGE DISCOVERY 2013年第3期26卷 452-511页

作者： Gullo, Francesco Domeniconi, Carlotta Tagarelli, Andrea Yahoo Res Barcelona 08018 Spain Univ Calabria DEIS Dept I-87036 Arcavacata Di Rende CS Italy George Mason Univ Dept Comp Sci Fairfax VA 22030 USA

A considerable amount of work has been done in data clustering research during the last four decades, and a myriad of methods has been proposed focusing on different data types, proximity functions, cluster representation models, and cluster presentation. However, clustering remains a challenging problem due to its ill-posed nature: it is well known that off-the-shelf clustering methods may discover different patterns in a given set of data, mainly because every clustering algorithm has its own bias resulting from the optimization of different criteria. This bias becomes even more important as in almost all real-world applications, data is inherently high-dimensional and multiple clustering solutions might be available for the same data collection. In this respect, the problems of projective clustering and clustering ensembles have been recently defined to deal with the high dimensionality and multiple clusterings issues, respectively. Nevertheless, despite such two issues can often be encountered together, existing approaches to the two problems have been developed independently of each other. In our earlier work Gullo et al. (Proceedings of the international conference on data mining (ICDM), 2009a) we introduced a novel clustering problem, called projective clustering ensembles (PCE): given a set (ensemble) of projective clustering solutions, the goal is to derive a projective consensus clustering, i.e., a projective clustering that complies with the information on object-to-cluster and the feature-to-cluster assignments given in the ensemble. In this paper, we enhance our previous study and provide theoretical and experimental insights into the PCE problem. PCE is formalized as an optimization problem and is designed to satisfy desirable requirements on independence from the specific clustering ensemble algorithm, ability to handle hard as well as soft data clustering, and different feature weightings. Two PCE formulations are defined: a two-objective optimizatio

关键词： clustering clustering ensembles Projective clustering Multi-objective optimization

来源：评论

学校读者我要写书评

暂无评论

Metacluster-based Projective clustering ensembles

引用

MACHINE LEARNING 2015年第1-2期98卷 181-216页

作者： Gullo, Francesco Domeniconi, Carlotta Tagarelli, Andrea Yahoo Res Barcelona 08018 Spain George Mason Univ Dept Comp Sci Fairfax VA 22030 USA Univ Calabria DIMES Dept I-87036 Arcavacata Di Rende CS Italy

The Projective clustering Ensemble (PCE) problem is a recent clustering advance aimed at combining the two powerful tools of clustering ensembles and projective clustering. PCE has been formalized as either a two-objective or a single-objective optimization problem. Two-objective PCE has been recognized as more accurate than its single-objective counterpart, although it is unable to jointly handle the object-based and feature-based cluster representations. In this paper, we push forward the current PCE research, aiming to overcome the limitations of all existing PCE formulations. We propose a novel single-objective PCE formulation so that (i) the object-based and feature-based cluster representations are jointly considered, and (ii) the resulting optimization strategy follows a metacluster-based methodology borrowed from traditional clustering ensembles. As a result, the proposed formulation features best suitability to the PCE problem, thus guaranteeing improved effectiveness. Experiments on benchmark datasets have shown how the proposed approach achieves better average accuracy than all existing PCE methods, as well as efficiency superior to the most accurate existing metacluster-based PCE method on larger datasets.

关键词： clustering clustering ensembles Projective clustering Subspace clustering Dimensionality reduction Optimization

来源：评论

学校读者我要写书评

暂无评论

Unsupervised feature selection using clustering ensembles and population based incremental learning algorithm

引用

PATTERN RECOGNITION 2008年第9期41卷 2742-2756页

作者： Hong, Yi Kwong, Sam Chang, Yuchou Ren, Qingsheng City Univ Hong Kong Dept Comp Sci Kowloon Hong Kong Peoples R China Brigham Young Univ Dept Elect & Comp Engn Provo UT 84602 USA Shanghai Jiao Tong Univ Dept Comp Sci & Engn Shanghai 200030 Peoples R China

This paper describes a novel feature selection algorithm for unsupervised clustering, that combines the clustering ensembles method and the population based incremental learning algorithm. The main idea of the proposed unsupervised feature selection algorithm is to search for a subset of all features such that the clustering algorithm trained on this feature subset can achieve the most similar clustering solution to the one obtained by an ensemble learning algorithm. In particular, a clustering solution is firstly achieved by a clustering ensembles method, then the population based incremental learning algorithm is adopted to find the feature subset that best fits the obtained clustering solution. One advantage of the proposed unsupervised feature selection algorithm is that it is dimensionality-unbiased. In addition, the proposed unsupervised feature selection algorithm leverages the consensus across multiple clustering solutions. Experimental results on several real data sets demonstrate that the proposed unsupervised feature selection algorithm is often able to obtain a better feature subset when compared with other existing unsupervised feature selection algorithms. (c) 2008 Elsevier Ltd. All rights reserved.

关键词： clustering ensembles dimensionality unbiased population based incremental learning algorithm unsupervised feature selection

来源：评论

学校读者我要写书评

暂无评论

Resampling-based selective clustering ensembles

引用

PATTERN RECOGNITION LETTERS 2009年第3期30卷 298-305页

作者： Hong, Yi Kwong, Sam Wang, Hanli Ren, Qjngsheng City Univ Hong Kong Dept Comp Sci Kowloon Hong Kong Peoples R China Shanghai Jiao Tong Univ Dept Comp Sci & Engn Shanghai Peoples R China

Traditional clustering ensembles methods combine all obtained clustering results at hand. However, we observe that it can often achieve a better clustering solution if only part of all available clustering results are combined. This paper proposes a novel clustering ensembles method, termed as resampling-based selective clustering ensembles method. The proposed selective clustering ensembles method works by evaluating the qualities of all obtained clustering results through resampling technique and selectively choosing part of promising clustering results to build the ensemble committee. The final solution is obtained through combining the clustering results of the ensemble committee. Experimental results on several real data sets demonstrate that resampling-based selective clustering ensembles method is often able to achieve a better solution when compared with traditional clustering ensembles methods. (C) 2008 Elsevier B.V. All rights reserved.

关键词： clustering analysis clustering ensembles Resampling technique

来源：评论

学校读者我要写书评

暂无评论

FedCE: Personalized Federated Learning Method based on clustering ensembles 23

FedCE: Personalized Federated Learning Method based on Clust...

引用

31st ACM International Conference on Multimedia (MM)

作者： Cai, Luxin Chen, Naiyue Cao, Yuanzhouhan He, Jiahuan Li, Yidong Beijing Jiaotong Univ Sch Comp & Informat Technol Minist Educ Key Lab Big Data & Artificial Intelligence Transp Beijing Peoples R China Beijing Jiaotong Univ Sch Comp & Informat Technol Beijing Peoples R China

ISBN: (纸本)9798400701085

Federated learning (FL) is a privacy-aware computing framework that enables multiple clients to collaborate in solving machine learning problems. In real scenarios, non-IID data held by different edge devices will degrade the performance of global FL models. To address this issue, most FL methods utilize cluster algorithms to group clients with similar distributions. However, these methods do not fully utilize the distribution features of client data, resulting in a lack of generalization in the cluster model. In order to make the cluster more suitable for the distribution features of user data, we propose a clustering-ensemble based federated learning method (FedCE) that sets each client associated with multiple clusters. We extract the features of client distributions to quantify the relationship between clients and clusters, and optimize the local model of clients through the historical performance of the cluster model. Furthermore, we dynamically estimate the number of clusters each client belongs to through the diversity of client performance. We conduct experiments on scenarios with mixture two-distributions, three-distributions and Dirichlet-distributions. The results show that the FedCE algorithm has better performance than the state-of-the-art clustered FL methods in both cluster and client models under different data distributions.

关键词： Federated Learning Personalized clustering ensembles

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：