Support Vector Number Reduction: Survey and Experimental Evaluations

Jung, Ho Gi; Kim, Gahyun

doi:10.1109/tits.2013.2282635

Cited by 20 publications

(15 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In particular, in the case of the third, fourth, sixth and seventh reduced‐set SV, most of their neighbouring SVs are from a single person. Although this confirms that a reduced‐set SV represents multiple SVs of the original SVM sharing common characteristics, it is suspicious that the SVNR performance of IPA in [3] might be exaggerated due to the multiple images of appearing persons in the database. Therefore, we naturally come to learn that experiments using databases without duplicated persons are required to unquestionably verify the SVNR performance of an IPA.…”

Section: Experimental Results Of Analysis Of Reduced‐set Constructionmentioning

confidence: 87%

“…This has been researched continuously since Burges et al introduced their reduced-set method in 1996 [2]. Jung et al published a survey and experimental evaluations concerning SVNR [3]. Their paper categorises SVNR methods into either pre-pruning or post-pruning according to whether they exploit the results of a standard SVM.…”

Section: Introductionmentioning

confidence: 99%

“…Experimental results with pedestrian and vehicle image databases confirm that the proposed method shows better performance than the HOGgles. By applying the proposed method to the resultant SVs of the IPA in [3] and comparing the reconstructed image with images of neighbouring SVs, it is found that the reconstructed image seems to be the composition of the images of neighbouring SVs. Furthermore, it is found that, in a large portion of the resultant SVs, the resultant SVs and their neighbouring SVs seem to correspond to a single person.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Analysis of reduced‐set construction using image reconstruction from a HOG feature vector

Jung

2017

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

Recently, several methods have been published that demonstrate how to reconstruct an image from a discriminative feature vector. This study explains that previous approaches minimising the histogram-of-oriented-gradient (HOG) feature error in the principal component analysis (PCA) domain of the learning database have a disadvantage in that they cannot reflect the different dynamic range of each PCA dimension, and proposes an improved method to exploit the eigenvalue as the weighting factor of each PCA dimension. Experimental results using pedestrian and vehicle image databases quantitatively show that the proposed method improves the quality of reconstructed images. Additionally, the proposed method is applied to the image reconstruction of the resultant support vectors (SVs) of reduced-set construction which showed the best performance among SV number reduction methods. As the resultant SVs of reduced-set construction are not corresponding to any image of the learning database, it is hard to analyse the problem and performance of the method. By observing the images of the resultant SVs, one potential problem regarding the database used is newly considered and the direction of further study can be established in order to address the problem.

show abstract

Section: Experimental Results Of Analysis Of Reduced‐set Constructionmentioning

confidence: 87%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Analysis of reduced‐set construction using image reconstruction from a HOG feature vector

Jung

2017

IET Computer Vision

Self Cite

View full text Add to dashboard Cite

show abstract

“…It is established on the structural risk minimization principle from the computational learning theory. The main aim of this principle is to find the hypothesis values at the lowest error rate Jung and Kim (2014). The linear kernel threshold function can be used in this research work.…”

Section: Methodsmentioning

confidence: 99%

Artificial bee colony algorithm for feature selection and improved support vector machine for text classification

Balakumar

Mohan

2019

IDD

View full text Add to dashboard Cite

Purpose Owing to the huge volume of documents available on the internet, text classification becomes a necessary task to handle these documents. To achieve optimal text classification results, feature selection, an important stage, is used to curtail the dimensionality of text documents by choosing suitable features. The main purpose of this research work is to classify the personal computer documents based on their content. Design/methodology/approach This paper proposes a new algorithm for feature selection based on artificial bee colony (ABCFS) to enhance the text classification accuracy. The proposed algorithm (ABCFS) is scrutinized with the real and benchmark data sets, which is contrary to the other existing feature selection approaches such as information gain and χ2 statistic. To justify the efficiency of the proposed algorithm, the support vector machine (SVM) and improved SVM classifier are used in this paper. Findings The experiment was conducted on real and benchmark data sets. The real data set was collected in the form of documents that were stored in the personal computer, and the benchmark data set was collected from Reuters and 20 Newsgroups corpus. The results prove the performance of the proposed feature selection algorithm by enhancing the text document classification accuracy. Originality/value This paper proposes a new ABCFS algorithm for feature selection, evaluates the efficiency of the ABCFS algorithm and improves the support vector machine. In this paper, the ABCFS algorithm is used to select the features from text (unstructured) documents. Although, there is no text feature selection algorithm in the existing work, the ABCFS algorithm is used to select the data (structured) features. The proposed algorithm will classify the documents automatically based on their content.

show abstract

“…Among the proposed techniques, that falls into the aforementioned categories, it is worth mentioning the usage of: (i) random selection techniques [6], (ii) clustering techniques [7] or graph-based techniques [8]. One can refer to thorough surveys that have been done recently by Garcia et al [9] in 2012 (for NN based classification), and by Jung et al [10] in 2014 for (Support Vector Machine (SVM) [11] based classification).…”

Section: Related Workmentioning

confidence: 99%

CNN Training with Graph-Based Sample Preselection: Application to Handwritten Character Recognition

Rayar

Goto

Uchida

2018

2018 13th IAPR International Workshop on Document Analysis Systems (DAS)

View full text Add to dashboard Cite

In this paper, we present a study on sample preselection in large training data set for CNN-based classification. To do so, we structure the input data set in a network representation, namely the Relative Neighbourhood Graph, and then extract some vectors of interest. The proposed preselection method is evaluated in the context of handwritten character recognition, by using two data sets, up to several hundred thousands of images. It is shown that the graph-based preselection can reduce the training data set without degrading the recognition accuracy of a non pretrained CNN shallow model.

show abstract

Support Vector Number Reduction: Survey and Experimental Evaluations

Cited by 20 publications

References 82 publications

Analysis of reduced‐set construction using image reconstruction from a HOG feature vector

Analysis of reduced‐set construction using image reconstruction from a HOG feature vector

Artificial bee colony algorithm for feature selection and improved support vector machine for text classification

CNN Training with Graph-Based Sample Preselection: Application to Handwritten Character Recognition

Contact Info

Product

Resources

About