k-nearest-neighbor Bayes-risk estimation

Fukunaga, Keinosuke; Hostetler, L.D.

doi:10.1109/tit.1975.1055373

Cited by 134 publications

(68 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Binary pattern classification is the quintessential pattern recognition problem studied at great length in the machine learning literature, with various algorithms such as the nearest neighbor rule (Cover & Hart, 1967;Fukunaga & Hostetler, 1975), support vector machines (Cortes & Vapnik, 1995;Vapnik, 1998), artificial neural networks (Haykin, 2008), and discriminant functions (McLachlan, 2004). These methods can be trained to construct decision rules to separate the pattern classes of interest based on a training set of data containing samples representing the two classes.…”

Section: The Supervised Learning Methods Used For Automatic Segmentationmentioning

confidence: 99%

Usul and Makam driven automatic melodic segmentation for Turkish music

Bozkurt

Karaosmanoğlu

Karaçali

et al. 2014

Journal of New Music Research

View full text Add to dashboard Cite

Automatic melodic segmentation is a topic studied extensively, aiming at developing systems that perform grouping of musical events. Here, we consider the problem of automatic segmentation via supervised learning from a dataset containing segmentation labels of an expert. We present a statistical classification-based segmentation system developed specifically for Turkish makam music. The proposed system uses two novel features, a makam-based and an usul-based feature, together with features commonly used in literature. The makam-based feature is defined as the probability of a note to appear at the phrase boundary, computed from the distributions of boundaries with respect to the piece's makam pitches. Likewise, the usul-based feature is computed from the distributions of boundaries with respect to beats in the rhythmic cycle, usul of the piece. Several experimental setups using different feature groups are designed to test the contribution of the proposed features on three datasets. The results show that the new features carry complementary information to existing features in the literature within the Turkish makam music segmentation context and that the inclusion of new features resulted in statistically significant performance improvement.

show abstract

Section: The Supervised Learning Methods Used For Automatic Segmentationmentioning

confidence: 99%

Usul and Makam driven automatic melodic segmentation for Turkish music

Bozkurt

Karaosmanoğlu

Karaçali

et al. 2014

Journal of New Music Research

View full text Add to dashboard Cite

show abstract

“…Later, he shown some of the formal properties of this procedure, for example, that the classification error rate is bounded by twice the Bayes error value when you have an infinite number of samples for classifying and k is equal to 1 (Cover & Hart, 1967). Once developed the formal properties of this classifier, he established a line of research that goes up today, highlighting the work of Hellman (Hellman, 1970), which show a new approach to rejection, Fukunaga and Hostetler (Fukunaga & Hostetler, 1975), which sets out refinements with respect to the Bayes error rate, or those developed by (Dudani, 1976) and Bailey and Jain (Bailey & Jain, 1978), in which new approaches were established to the use of weighted distances. Other interesting work on the subject is related to soft computing (Bermejo & Cabestany, 2000) and fuzzy methods (Jozwik, 1983, Keller et al, 1985.…”

Section: K Nearest Neighbors (Knn)mentioning

confidence: 99%

Thermooxidative Properties of Biodiesels and Other Biological Fuels

Tarrío-Saavedra¹,

Naya²,

López-Beceiro³

et al. 2011

Biodiesel- Quality, Emissions and by-Products

View full text Add to dashboard Cite

“…where pðxjx A C 0 Þ and pðxjx A C 1 Þ represent the class conditional probability densities for the classes C 0 and C 1 , respectively [4,10]. The basis of this observation is as follows: Let N ðxÞ & X be a small spherical neighborhood of x of size VðN ðxÞÞ, and R be a random reference set containing n points from each of C 0 and C 1 .…”

Section: Likelihood Ratio Estimation Via the Nearest Neighbor Rulementioning

confidence: 99%

“…The general strategy relies on expert-curated ground truth datasets providing the categorical associations of all available data samples. Ground truth datasets are then used as the basis for statistical learning, specifically to construct a classification rule using one of a host of methods such as the support vector machines [1,2], nearest neighbor classifiers [3,4], neural networks [5], discriminant functions [6] and so on.…”

Section: Introductionmentioning

confidence: 99%

Quasi-supervised learning for biomedical data analysis

Karaçali

2010

Pattern Recognition

View full text Add to dashboard Cite

a b s t r a c tWe present a novel formulation for pattern recognition in biomedical data. We adopt a binary recognition scenario where a control dataset contains samples of one class only, while a mixed dataset contains an unlabeled collection of samples from both classes. The mixed dataset samples that belong to the second class are identified by estimating posterior probabilities of samples for being in the control or the mixed datasets. Experiments on synthetic data established a better detection performance against possible alternatives. The fitness of the method in biomedical data analysis was further demonstrated on real multi-color flow cytometry and multi-channel electroencephalography data.

show abstract

k-nearest-neighbor Bayes-risk estimation

Cited by 134 publications

References 11 publications

Usul and Makam driven automatic melodic segmentation for Turkish music

Usul and Makam driven automatic melodic segmentation for Turkish music

Thermooxidative Properties of Biodiesels and Other Biological Fuels

Quasi-supervised learning for biomedical data analysis

Contact Info

Product

Resources

About