Scholarlink

This study aims to analyze multidimensional data using various machine learning methods such as Kohonen maps, decision trees, and neural networks. The introduction describes the importance of solving the problem of analyzing complex multidimensional data, and also highlights the goals of the study. Kohonen maps were used to visualize the feature space and identify structures in the data. Decision trees helped to identify key features and support the decision-making process, and neural networks were successfully applied to highly accurate prediction of the target variable. In the section “Results” and “Discussion”, a comparison is made with additional existing models, which emphasizes the effectiveness of the proposed methods. The paper also considers hybrid models combining Kohonen maps, decision trees and neural networks to improve prediction accuracy. Figures 2 and 3 are explained in detail, which allows for a better understanding of the results obtained. In conclusion, quantitative results and indicators are presented that confirm the claims of effectiveness, such as the accuracy of the neural network, key characteristics identified using decision trees, and structures discovered using Kohonen maps. The proposed approach can be applied in various fields to analyze and predict complex phenomena and processes.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Analysis of Multivariate Data Using Machine Learning Techniques Such as Self-organizing Kohonen Maps, Decision Tree and Neural Network

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Vladislav Kukartsev,
  • Ksenia Degtyareva,
  • Ekaterina Volneikina,
  • Alena Rozhkova

摘要

This study aims to analyze multidimensional data using various machine learning methods such as Kohonen maps, decision trees, and neural networks. The introduction describes the importance of solving the problem of analyzing complex multidimensional data, and also highlights the goals of the study. Kohonen maps were used to visualize the feature space and identify structures in the data. Decision trees helped to identify key features and support the decision-making process, and neural networks were successfully applied to highly accurate prediction of the target variable. In the section “Results” and “Discussion”, a comparison is made with additional existing models, which emphasizes the effectiveness of the proposed methods. The paper also considers hybrid models combining Kohonen maps, decision trees and neural networks to improve prediction accuracy. Figures 2 and 3 are explained in detail, which allows for a better understanding of the results obtained. In conclusion, quantitative results and indicators are presented that confirm the claims of effectiveness, such as the accuracy of the neural network, key characteristics identified using decision trees, and structures discovered using Kohonen maps. The proposed approach can be applied in various fields to analyze and predict complex phenomena and processes.