A Hybrid Feature Extraction Selection Approach for High-Dimensional Non-Gaussian Data Clustering

Boutemedjet, Sabri; Bouguila, Nizar; Ziou, Djemel

doi:10.1109/tpami.2008.155

Cited by 118 publications

(88 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…RULe_GDM_FS and the method in [113]. As expected , noise affects the parameters estimated with the EM method.…”

Section: (D)supporting

confidence: 55%

“…Each cluster contains 3000 points. We increase the noise rate Table 3.3: Comparaison of the parameters used to generate the data to the parameters estimated using the method in [113] and RULe_ GDM_FS.…”

Section: (D)mentioning

confidence: 99%

“…We compare our proposed image annotation approach to three state-of-the-art v methods and demonstrate the effectiveness of the proposed image retrieval system. (a) data set, (b) partition obtained with the method in [113], (c) partition obtained using Gaussian mixture model as described in [85], and (d) partition obtained using RULe_ GDM_FS.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Image annotation and retrieval based on multi-modal feature clustering and similarity propagation.

Ismail¹,

Maher²

View full text Add to dashboard Cite

“…RULe_GDM_FS and the method in [113]. As expected , noise affects the parameters estimated with the EM method.…”

Section: (D)supporting

confidence: 55%

Section: (D)mentioning

confidence: 99%

See 1 more Smart Citation

Image annotation and retrieval based on multi-modal feature clustering and similarity propagation.

Ismail¹,

Maher²

View full text Add to dashboard Cite

“…The redundancy present in the object's data only slightly improves classifier performance [1]. Besides, time and space requirements will increase significantly when the data contains a large number of redundancy features.…”

Section: Introductionmentioning

confidence: 99%

Local Reconstruction Error Alignment: A Fast Unsupervised Feature Selection Algorithm for Radar Target Clustering

Wang

Chen

2014

IEICE Trans. Inf. & Syst.

View full text Add to dashboard Cite

SUMMARYObserved samples in wideband radar are always represented as nonlinear points in high dimensional space. In this paper, we consider the feature selection problem in the scenario of wideband radar target clustering. Inspired by manifold learning, we propose a novel feature selection algorithm, called Local Reconstruction Error Alignment (LREA), to select the features that can best preserve the underlying manifold structure. We first select the features that minimize the reconstruction error in every neighborhood. Then, we apply the alignment technique to extend the local optimal feature sequence to a global unique feature sequence. Experiments demonstrate the effectiveness of our proposed method.

show abstract

“…Automatic feature extraction methods for distribution-based clustering have also been proposed, e.g. in [11].…”

Section: Introductionmentioning

confidence: 99%

Compressive Clustering of High-Dimensional Data

Ruta

Porikli

2012

2012 11th International Conference on Machine Learning and Applications

View full text Add to dashboard Cite

In this paper we focus on realistic clustering problems where the input data is high-dimensional and the clusters have complex, multimodal distribution. In this challenging setting the conventional methods, such as k-centers family, hierarchical clustering or those based on model fitting, are inefficient and typically converge far from the globally optimal solution. As an alternative, we propose a novel unsupervised learning approach which is based on the compressive sensing paradigm. The key idea underlying our algorithm is to monitor the distance between the test sample and its principal projection in each cluster, and continue re-assigning it to the cluster yielding the smallest residual. As a result, we obtain an iterative procedure which, under the compressive assumptions, minimizes the total reconstruction error of all samples from their nearest clusters. To evaluate the proposed approach, we have conducted a series of experiments involving various image collections where the task was to automatically group similar objects. Comparison of the obtained results with those yielded by the state-of-the-art clustering methods provides evidence for high discriminative power of our algorithm. International Conference on Machine Learning and Applications (ICMLA)This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. Abstract-In this paper we focus on realistic clustering problems where the input data is high-dimensional and the clusters have complex, multimodal distribution. In this challenging setting the conventional methods, such as k-centers family, hierarchical clustering or those based on model fitting, are inefficient and typically converge far from the globally optimal solution. As an alternative, we propose a novel unsupervised learning approach which is based on the compressive sensing paradigm. The key idea underlying our algorithm is to monitor the distance between the test sample and its principal projection in each cluster, and continue re-assigning it to the cluster yielding the smallest residual. As a result, we obtain an iterative procedure which, under the compressive assumptions, minimizes the total reconstruction error of all samples from their nearest clusters. To evaluate the proposed approach, we have conducted a series of experiments involving various image collections where the task was to automatically group similar objects. Comparison of the obtained results with thos...

show abstract

A Hybrid Feature Extraction Selection Approach for High-Dimensional Non-Gaussian Data Clustering

Cited by 118 publications

References 30 publications

Image annotation and retrieval based on multi-modal feature clustering and similarity propagation.

Image annotation and retrieval based on multi-modal feature clustering and similarity propagation.

Local Reconstruction Error Alignment: A Fast Unsupervised Feature Selection Algorithm for Radar Target Clustering

Compressive Clustering of High-Dimensional Data

Contact Info

Product

Resources

About