Stochastic Game Model of Data Clustering
Authors:
- Petro KravetsPKPetro Kravets
- Yevhen BurovYBYevhen Burov
- Oksana OborskaOOOksana Oborska
- Victoria VysotskaVVVictoria Vysotska
- Lyudmyla DzyubykLDLyudmyla Dzyubyk
- Vasyl LytvynVLVasyl Lytvyn
- 23 April 2021
- report
- Published by Academy of Cognitive and Natural Sciences
Abstract
A stochastic game model of data clustering under interference conditions is proposed. An adaptive recurrent method and algorithm for stochastic game deciding have developed. Computer simulation of game clustering of noisy data has performed. The parameters influence on the stochastic game method convergence for noisy data clustering is researched. For this purpose, each data point is considered as a separate player with the ability to learn and adapt to the uncertainties of the system. After the selection of clusters is completed, all players calculate the corresponding losses by the criteria of minimizing the total distance between the cluster points formed by the free choice of player strategies. The results obtained are analysed. A stochastic approximation based on the mixed-strategy adjustment method minimizes the mean loss functions on single simplexes.Keywords
This publication has 24 references indexed in Scilit:
- Forest Cover Type Classification Based on Environment Characteristics and Machine Learning Technology2020
- Detecting Items with the Biggest Weight Based on Neural Network and Machine Learning Methods2020
- Approach for Modeling Search Web-Services Based on Color Petri Nets2020
- Methods for Forecasting Nonlinear Non-stationary Processes in Machine Learning2020
- Adaptive Strategies in the Multi-agent “Predator-Prey” Models2020
- Dynamic Coordination of Strategies for Multi-agent Systems2020
- An Evaluation of the Objective Clustering Inductive Technology Effectiveness Implemented Using Density-Based and Agglomerative Hierarchical Clustering Algorithms2019
- A hybrid approach using genetic algorithm and the differential evolution heuristic for enhanced initialization of the k-means algorithm with applications in text clusteringSoft Computing, 2018
- Integration of artificial immune network and K-means for cluster analysisKnowledge and Information Systems, 2013
- Game Theory2013