Abstract:Polymer materials have garnered significant attention due to their exceptional mechanical properties and diverse industrial applications. Understanding the glass transition temperature (Tg) of polymers is critical to prevent operational failures at specific temperatures. Traditional methods for measuring Tg, such as differential scanning calorimetry (DSC) and dynamic mechanical analysis, while accurate, are often time-consuming, costly, and susceptible to inaccuracies due to random and uncertain factors. To address these limitations, the aim of the present study is to investigate the potential of Simplified Molecular Input Line Entry System (SMILES) as descriptors in simple machine learning models to predict Tg efficiently and reliably. Five models were utilized: k-nearest neighbors (KNNs), support vector regression (SVR), extreme gradient boosting (XGBoost), artificial neural network (ANN), and recurrent neural network (RNN). SMILES descriptors were converted into numerical data using either One Hot Encoding (OHE) or Natural Language Processing (NLP). The study found that SMILES inputs with fewer than 200 characters were inadequate for accurately describing compound structures, while inputs exceeding 200 characters diminished model performance due to the curse of dimensionality. The ANN model achieved the highest R2 value of 0.79; however, the XGB model, with an R2 value of 0.774, exhibited the highest stability and shorter training times compared to other models, making it the preferred choice for Tg prediction. The efficiency of the OHE method over NLP was demonstrated by faster training times across the KNN, SVR, XGB, and ANN models. Validation of new polymer data showed the XGB model's robustness, with an average prediction deviation of 9.76 from actual Tg values. These findings underscore the importance of optimizing SMILES conversion methods and model parameters to enhance prediction reliability. Future research should focus on improving model accuracy and generalizability by incorporating additional features and advanced techniques. This study contributes to the development of efficient and reliable predictive models for polymer properties, facilitating the design and application of new polymer materials.

Prediction of the Glass Transition Temperatures for Polymers with Artificial Neural Network

A neural network approach to prediction of glass transition temperature of polymers

Predicting Chain Dimensions From An Artificial Neural Network Model

Prediction of Glass Transition Temperature of Polymer by Support Vector Regression

Modeling and Predicting the Glass Transition Temperature of Vinyl Polymers by Using Hybrid Pso-Svr Method

Prediction of the Glass Transition Temperatures of Styrenic Copolymers by Using Support Vector Regression Combined with Particle Swarm Optimization

Predicting Glass Transition Temperature of Polymers by Combining Molecular Dynamics Simulations and Machine Learning Techniques

Efficient Prediction of Polymer Glass Transition Temperatures through Machine Learning Methods

A glimpse inside materials: Polymer structure – Glass transition temperature relationship as observed by a trained artificial intelligence

Prediction of Glass Transition Temperature of Polymers Using Simple Machine Learning

Modeling and Predicting the Glass Transition Temperature of Polymethacrylates Based on Quantum Chemical Descriptors by Using Hybrid PSO‐SVR

Artificial Neural Network Prediction of Steric Hindrance Parameter of Polymers

Machine learning analysis of a large set of homopolymers to predict glass transition temperatures

Predicting glass transition of amorphous polymers by application of cheminformatics and molecular dynamics simulations

Molecular Dynamics Simulation on the Glass Transition Temperatures of TPI/NR/LDPE

Novel descriptors from main and side chains of high-molecular-weight polymers applied to prediction of glass transition temperatures

Accurate prediction of θ (lower critical solution temperature) in polymer solutions based on 3D descriptors and artificial neural networks

Modeling the Glass Transition Temperature of Polymers Via Multipole Moments Using Support Vector Regression

Large-Scale Glass-Transition Temperature Prediction with an Equivariant Neural Network for Screening Polymers

Correlation Between the Glass Transition Temperatures and Multipole Moments for Polymers

A fast transferable method for predicting the glass transition temperature of polymers from chemical structure