Solve Classification Tasks with Probabilities. Statistically-Modeled Outputs
Abstract
In This Paper, an Approach for Probability-Based Class Prediction is Presented. This Approach is based on a Combination of a Newly Proposed Histogram Probability (Hp) Method and Any Classification Algorithm (In This Paper Results for Combination with Extreme Learning Machines (Elm) and Support Vector Machines (Svm) Are Presented). Extreme Learning Machines is a Method of Training a Single-Hidden Layer Neural Network. the Paper Contains Detailed Description and Analysis of the Hp Method by the Example of the Iris Dataset. Eight Datasets, Four of Which Represent Computer Vision Classification Problem and Are Derived from Caltech-256 Image Database, Are Used to Compare Hp Method with Another Probability-Output Classifier [11, 18].
Recommended Citation
A. Gritsenko et al., "Solve Classification Tasks with Probabilities. Statistically-Modeled Outputs," Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10334 LNCS, pp. 293 - 305, Springer, Jan 2017.
The definitive version is available at https://doi.org/10.1007/978-3-319-59650-1_25
Department(s)
Engineering Management and Systems Engineering
Keywords and Phrases
Classification; Extreme learning machines; Gaussian mixture model; Histogram distribution; Image recognition; Machine learning; Multiclass classification; Probabilistic classification
International Standard Book Number (ISBN)
978-331959649-5
International Standard Serial Number (ISSN)
1611-3349; 0302-9743
Document Type
Article - Conference proceedings
Document Version
Citation
File Type
text
Language(s)
English
Rights
© 2024 Springer, All rights reserved.
Publication Date
01 Jan 2017
