Application of machine learning method for the classification of carbon and fluorocarbon films

The article describes a method for the classification of carbon and fluorocarbon films using machine learning algorithms. The method of support vectors was chosen as the main classification algorithm. The kernel of the method was the radial Gauss basis function. Thus, the algorithm was presented as...

Full description

Saved in:
Bibliographic Details
Main Authors: Uvarov, Sergey, Uvarova, Olga, Shchur, Pavel
Format: Conference Proceeding
Language:English
Subjects:
Algorithms
Artificial intelligence
Basis functions
Carbon
Classification
Computer architecture
Machine learning
Neural networks
Perfluorocarbons
Software
Training
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The article describes a method for the classification of carbon and fluorocarbon films using machine learning algorithms. The method of support vectors was chosen as the main classification algorithm. The kernel of the method was the radial Gauss basis function. Thus, the algorithm was presented as a two-layer neural network of direct signal propagation, where intermediate layer contains radially symmetric neurons. For the neural network training, data provided by the experimenters were used. The paper describes the process of neural network training and features of the software implementation of distributed algorithm based on client-server architecture. The constructed neural network allows us to classify films according to the anti-adhesive class. To achieve high performance at the training stage, graphic processes were used. In addition, high portability of the developed software was provided thanks to virtualization methods.
ISSN:0094-243X
1551-7616
DOI:10.1063/1.5135680