NUMERICAL METHODS AND DATA ANALYSIS
Multivariate mixed kernel density estimators and their application in machine learning for classification of biological objects based on spectral measurements
A. A. Sirota, A. O. Donskikh, A. V. Akimov, D. A. Minakov
Voronezh State University, Voronezh, Russia
A problem of non-parametric multivariate density estimation for machine learning and data augmentation is considered. A new mixed density estimation method based on calculating the convolution of independently obtained kernel density estimates for unknown distributions of informative features and a known (or independently estimated) density for non-informative interference occurring during measurements is proposed. Properties of the mixed density estimates obtained using this method are analyzed. The method is compared with a conventional ParzenRosenblatt window method applied directly to the training data. The equivalence of the mixed kernel density estimator and the data augmentation procedure based on the known (or estimated) statistical model of interference is theoretically and experimentally proven. The applicability of the mixed density estimators for training of machine learning algorithms for the classification of biological objects (elements of grain mixtures) based on spectral measurements in the visible and near-infrared regions is evaluated.
machine learning, pattern classification, data augmentation, kernel density estimation, spectral measurements.
|Ministry of Science and Higher Education of the Russian Federation
|The presented study was supported by the Ministry of Education and Science of the Russian Federation under project No. 8.3844.2017/4.6 ("Development of facilities for express analysis and classification of the components of nonuniform grain mixtures with pathologies based on the integration between spectral analysis methods and machine learning").
PDF file (1317 kB)
A. A. Sirota, A. O. Donskikh, A. V. Akimov, D. A. Minakov, “Multivariate mixed kernel density estimators and their application in machine learning for classification of biological objects based on spectral measurements”, Computer Optics, 43:4 (2019), 677–691
Citation in format AMSBIB
\by A.~A.~Sirota, A.~O.~Donskikh, A.~V.~Akimov, D.~A.~Minakov
\paper Multivariate mixed kernel density estimators and their application in machine learning for classification of biological objects based on spectral measurements
\jour Computer Optics
Citing articles on Google Scholar:
Related articles on Google Scholar:
|Number of views:|