Feature selection using structural similarity

Kundu, Partha Pratim; Pedrycz, Witold

doi:10.1016/j.ins.2012.02.042

Cited by 31 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sanguinetti [39] presented a latent variable model to perform dimensionality reduction on a dataset that contained clusters; specifically, a variable was considered salient when it preserved clustered information by mapping an original representation space to a latent space. Mitra [32] proposed using structural similarity between clusters for feature selection, where the topological neighborhood information about pairs of instances was considered to assess the similarity. Furthermore, a feature selection method based on the wrapper model in semi-supervised feature selection, considering labeled and unlabeled examples, has also been described [50,53].…”

Section: Related Workmentioning

confidence: 99%

Feature selection for clustering using instance-based learning by exploring the nearest and farthest neighbors

Chen

2015

Information Sciences

View full text Add to dashboard Cite

Section: Related Workmentioning

confidence: 99%

Feature selection for clustering using instance-based learning by exploring the nearest and farthest neighbors

Chen

2015

Information Sciences

View full text Add to dashboard Cite

“…An approach based on cooperative game theory evaluates the power of each feature individually and within groups [32]. The structural similarity between data before and after feature selection is maintained and topological neighborhood information is used for computing the structural similarity [33]. An unsupervised feature ranking algorithm can discover Bi-clusters that are used to evaluate feature inter-dependencies, separability of instances and feature ranking [23].…”

Section: A Feature Selectionmentioning

confidence: 99%

Person Re-Identification

2014

Advances in Computer Vision and Pattern Recognition

236

View full text Add to dashboard Cite

Abstract-Networks of smart cameras share large amounts of data to accomplish tasks such as re-identification. We propose a feature selection method that minimizes the data needed to represent the appearance of objects by learning the most appropriate feature set for the task at hand (person re-identification). The computational cost for feature extraction and the cost for storing the feature descriptor are considered jointly with feature performance in order to select cost-effective good features. This selection allows us to improve inter-camera re-identification while reducing the bandwidth that is necessary to share data across the camera network. We also rank the selected features in the order of effectiveness for the task to enable a further reduction of the feature set by dropping the least effective features when application constraints require this adaptation. We compare the proposed approach with state-of-the-art methods on the i-LIDS and VIPeR datasets and show that the proposed approach considerably reduces network traffic due to inter-camera feature sharing while keeping the re-identification performance at an equivalent or better level compared with the state of the art.

show abstract

“…Since finding the best feature subset is in exponential space, the feature selection problem is intractable or NP-hard [2]. In order to overcome the intractable property of the feature selection problem, good search algorithms are required.…”

Section: Introductionmentioning

confidence: 99%

A hybrid search method of wrapper feature selection by chaos particle swarm optimization and local search

Javidi¹,

Emami²

2016

Turk J Elec Eng & Comp Sci

View full text Add to dashboard Cite

Finding a subset of features from a large dataset is a problem that arises in many fields of study. Since the increasing number of features has extended the computational cost of a system, it is necessary to design and implement a system with the least number of features. The purpose of feature selection is to find the best subset of features from the original ones. The result of the best selection is improving the computational cost and the accuracy of the prediction.A large number of algorithms have been proposed for feature subset selection. In this paper, we propose a wrapper feature selection algorithm for a classification that is based on chaos theory, binary particle swarm optimization, and local search. In the proposed algorithm, the nearest neighbor algorithm is used for the evaluation phase.

show abstract

Feature selection using structural similarity

Cited by 31 publications

References 23 publications

Feature selection for clustering using instance-based learning by exploring the nearest and farthest neighbors

Feature selection for clustering using instance-based learning by exploring the nearest and farthest neighbors

Person Re-Identification

A hybrid search method of wrapper feature selection by chaos particle swarm optimization and local search

Contact Info

Product

Resources

About