A Kernel Based Neighborhood Discriminant Submanifold Learning for Pattern Classification

We propose a novel method, called Kernel Neighborhood Discriminant Analysis (KNDA), which can be regarded as a supervised kernel extension of Locality Preserving Projection (LPP). KNDA nonlinearly maps the original data into a kernel space in which two graphs are constructed to depict the within-class submanifold and the between-class submanifold. Then a criterion function which minimizes the quotient between the within-class representation and the between-class representation of the submanifolds is designed to separate each submanifold constructed by each class. The real contribution of this paper is that we bring and extend the submanifold based algorithm to a general model and by some derivation a simple result is given by which we can classify a given object to a predefined class effectively. Experiments on the MNIST Handwritten Digits database, the Binary Alphadigits database, the ORL face database, the Extended Yale Face Database B, and a downloaded documents dataset demonstrate the effectiveness and robustness of the proposed method.

Download Full-text

Application of multivariate techniques in the evaluation of pure lines of beans

Ciência Rural ◽

10.1590/0103-8478cr20140329 ◽

2016 ◽

Vol 46 (9) ◽

pp. 1535-1541 ◽

Cited By ~ 3

Author(s):

Rodolfo Schmit ◽

Rita Carolina de Melo ◽

Thayse Cristine Vieira Pereira ◽

Mattheus Beck ◽

Altamir Frederico Guidolin ◽

...

Keyword(s):

Discriminant Analysis ◽

Discriminant Function ◽

Block Design ◽

Original Data ◽

Inbred Lines ◽

Canonical Discriminant Analysis ◽

Multivariate Techniques ◽

Testing Hypotheses ◽

Randomized Block Design ◽

Pure Lines

ABSTRACT: The objective of this study was to apply multivariate techniques, canonical discriminant analysis, and multivariate contrasts, indicating the most favorable inferences in the evaluation of pure lines of beans. The study was conducted at the experimental field of the Institute for Breeding and Molecular Genetics, in Lages, SC, Brazil. The experiment was composed of 24 pure lines of beans from the Santa Catarina test of cultivars. Plant height, numbers of pods and grains per plant, and stem diameter were the variables measured. The complete randomized block design was used with four replications. The data were subjected to multivariate analysis of variance, canonical discriminant analysis, multivariate contrasts and univariate contrasts. The first canonical discriminant function has captured 81% of the total variation in the data. The Scott-Knott test showed two groups of inbred lines at the average -of scores of the first canonical discriminant function. It was considered that testing hypotheses with the canonical scores may result in loss of information obtained from the original data. Multivariate contrasts indicated differences within the group formed by the Scott-Knott test. The canonical discriminant analysis and multivariate contrasts are excellent techniques to be combined in the multivariate assessment, being used to explore and test hypotheses, respectively.

Download Full-text

Embedding Undersampling Rotation Forest for Imbalanced Problem

Computational Intelligence and Neuroscience ◽

10.1155/2018/6798042 ◽

2018 ◽

Vol 2018 ◽

pp. 1-15 ◽

Cited By ~ 3

Author(s):

Huaping Guo ◽

Xiaoyu Diao ◽

Hongbing Liu

Keyword(s):

Imbalanced Data ◽

Feature Space ◽

Original Data ◽

Training Set ◽

Data Set ◽

Minority Class ◽

Rotation Forest ◽

Novel Method ◽

Individual Classifier ◽

The Cost

Rotation Forest is an ensemble learning approach achieving better performance comparing to Bagging and Boosting through building accurate and diverse classifiers using rotated feature space. However, like other conventional classifiers, Rotation Forest does not work well on the imbalanced data which are characterized as having much less examples of one class (minority class) than the other (majority class), and the cost of misclassifying minority class examples is often much more expensive than the contrary cases. This paper proposes a novel method called Embedding Undersampling Rotation Forest (EURF) to handle this problem (1) sampling subsets from the majority class and learning a projection matrix from each subset and (2) obtaining training sets by projecting re-undersampling subsets of the original data set to new spaces defined by the matrices and constructing an individual classifier from each training set. For the first method, undersampling is to force the rotation matrix to better capture the features of the minority class without harming the diversity between individual classifiers. With respect to the second method, the undersampling technique aims to improve the performance of individual classifiers on the minority class. The experimental results show that EURF achieves significantly better performance comparing to other state-of-the-art methods.

Download Full-text

Novel method for bearing performance degradation assessment – A kernel locality preserving projection-based approach

Proceedings of the Institution of Mechanical Engineers Part C Journal of Mechanical Engineering Science ◽

10.1177/0954406213486735 ◽

2013 ◽

Vol 228 (3) ◽

pp. 548-560 ◽

Cited By ~ 18

Author(s):

Chuang Sun ◽

Zhousuo Zhang ◽

Zhengjia He ◽

Zhongjie Shen ◽

Binqiang Chen ◽

...

Keyword(s):

Test Data ◽

Performance Degradation ◽

Inner Product ◽

Data Sets ◽

Degradation Index ◽

Locality Preserving Projection ◽

Locality Preserving ◽

Non Linear ◽

Linear Information ◽

Novel Method

Bearing performance degradation assessment is meaningful for keeping mechanical reliability and safety. For this purpose, a novel method based on kernel locality preserving projection is proposed in this article. Kernel locality preserving projection extends the traditional locality preserving projection into the non-linear form by using a kernel function and it is more appropriate to explore the non-linear information hidden in the data sets. Considering this point, the kernel locality preserving projection is used to generate a non-linear subspace from the normal bearing data. The test data are then projected onto the subspace to obtain an index for assessing bearing degradation degrees. The degradation index that is expressed in the form of inner product indicates similarity of the normal data and the test data. Validations by using monitoring data from two experiments show the effectiveness of the proposed method.

Download Full-text