Scholarlink

Abstract <p>In recent years, machine learning algorithms have become popular for predicting the physicochemical properties of polymers, including the glass transition temperature (<i>T</i><sub>g</sub>). Accurate <i>T</i><sub>g</sub> prediction is critical for developing polymers with desired properties. Traditional <i>T</i><sub>g</sub> prediction was based on semi-empirical methods, such as Askadskii’s method. The goal of this study was to develop a hybrid approach for predicting the <i>T</i><sub>g</sub> of organic homopolymers, combining Askadskii’s method and the QSPR model with machine learning (ML), which uses the advantages of theoretical analysis and the capabilities of ML to improve prediction accuracy. Random Forest, K-Nearest Neighbors, and a multilayer perceptron were used. The molecular structure of the polymers was represented by structural keys (MACCSKeys) and Morgan fingerprints. Optimization of the random forest algorithm hyperparameters enabled an <i>R</i><sup>2</sup> of up to 0.77 to be achieved on the test set. A comparative analysis showed that Morgan fingerprints that consider the spatial arrangement of fragments provide higher prediction accuracy, especially for isomeric homopolymers, where the spatial arrangement of substituents is important. The results demonstrate the potential of using ML for predicting polymer <i>T</i><sub>g</sub> based on glass transition theories and highlight the need for further research into hybrid models.</p>

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

A Hybrid Approach to Predicting Glass Transition Temperatures of Organic Homopolymers: A Combination of the QSPR Model and the Increment Method

  • 期刊论文

  • 发表日期：

    2025年8月1日

查看全文
  • G. R. Shadrina,
  • V. I. Anisimova,
  • I. S. Rodionov,
  • A. A. Baldinov,
  • N. V. Ulitin,
  • Ya. L. Lyulinskaya,
  • K. A. Tereshchenko,
  • D. A. Shiyan

摘要

Abstract

In recent years, machine learning algorithms have become popular for predicting the physicochemical properties of polymers, including the glass transition temperature (Tg). Accurate Tg prediction is critical for developing polymers with desired properties. Traditional Tg prediction was based on semi-empirical methods, such as Askadskii’s method. The goal of this study was to develop a hybrid approach for predicting the Tg of organic homopolymers, combining Askadskii’s method and the QSPR model with machine learning (ML), which uses the advantages of theoretical analysis and the capabilities of ML to improve prediction accuracy. Random Forest, K-Nearest Neighbors, and a multilayer perceptron were used. The molecular structure of the polymers was represented by structural keys (MACCSKeys) and Morgan fingerprints. Optimization of the random forest algorithm hyperparameters enabled an R2 of up to 0.77 to be achieved on the test set. A comparative analysis showed that Morgan fingerprints that consider the spatial arrangement of fragments provide higher prediction accuracy, especially for isomeric homopolymers, where the spatial arrangement of substituents is important. The results demonstrate the potential of using ML for predicting polymer Tg based on glass transition theories and highlight the need for further research into hybrid models.