Minimum spanning tree based one-class classifier

Juszczak, Piotr; Tax, David M. J.; Pekalska, Elbieta; Duin, Robert P. W.

doi:10.1016/j.neucom.2008.05.003

Cited by 122 publications

(80 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The details of the datasets are given in Table 1. The proposed method is compared to that described by Kim, et al [6] and one-class classifiers, such as Gaussian Data Description (GAUSSDD) [12], k-Means Data Description (KMEANSDD) [12], Minimum Spanning Tree Data Description (MSTDD) [13], k-Nearest Neighbor Data Description (KNNDD) [14], Support Vector Data Description (SVDD) [15], Linear Programming Data Description (LPDD) [16] and Minimax Probability Machine (MPMDD) [17]. For the method [6], we use the word score that is the result of finding the best match for the outlier detection, and SEG_T is assigned to integer numbers in the range [5], [15].…”

Section: Methodsmentioning

confidence: 99%

Improved Lexicon-driven based Chord Symbol Recognition in Musical Images

Dinh¹,

Ngoc²,

Yang³

et al. 2016

International Journal of Contents

View full text Add to dashboard Cite

show abstract

Section: Methodsmentioning

confidence: 99%

Improved Lexicon-driven based Chord Symbol Recognition in Musical Images

Dinh¹,

Ngoc²,

Yang³

et al. 2016

International Journal of Contents

View full text Add to dashboard Cite

show abstract

“…Gaussian model (Gauss), Mixture of Gaussians (MoG), Parzen Density Estimation (PDE) have been chosen as density methods. The boundary methods selected are k-centers (kC), k-nearest neighbor (kNN), k-means (kM) [30], Minimum Spanning Trees (MST) [31] and SVDD. 3 These methods are the most commonly used in the literature for one-class classification problems.…”

Section: Validation Methodologymentioning

confidence: 99%

Approximate polytope ensemble for one-class classification

Casale

Pujol

Radeva

2014

Pattern Recognition

View full text Add to dashboard Cite

a b s t r a c tIn this work, a new one-class classification ensemble strategy called approximate polytope ensemble is presented. The main contribution of the paper is threefold. First, the geometrical concept of convex hull is used to define the boundary of the target class defining the problem. Expansions and contractions of this geometrical structure are introduced in order to avoid over-fitting. Second, the decision whether a point belongs to the convex hull model in high dimensional spaces is approximated by means of random projections and an ensemble decision process. Finally, a tiling strategy is proposed in order to model nonconvex structures. Experimental results show that the proposed strategy is significantly better than state of the art one-class classification methods on over 200 datasets.

show abstract

“…Estimating the complete density or structure of a target concept in a one-class problem can very often be too demanding or even impossible. Boundary methods instead concentrate on estimating only the closed boundary for the given data, assuming that such a boundary will sufficiently describe the target class [28].…”

Section: Learning In the Absence Of Counterexamplesmentioning

confidence: 99%