A Comparative Study of Machine Learning Techniques for Predicting the Performance of Polymer Nanocomposites Incorporating Bio-Based Fillers

EasyChair Preprint 14516

Abstract

The performance of polymer nanocomposites can be significantly influenced by the incorporation of bio-based fillers, which offer both environmental and mechanical advantages. This study presents a comparative analysis of various machine learning techniques for predicting the performance of these advanced materials. We evaluate the efficacy of multiple algorithms, including linear regression, support vector machines (SVM), random forests, and deep neural networks, in forecasting key properties such as tensile strength, thermal stability, and elasticity of polymer nanocomposites containing bio-based fillers. Using a comprehensive dataset compiled from experimental studies and simulation data, we assess the predictive accuracy, computational efficiency, and generalization capabilities of each technique. The results reveal the strengths and limitations of each approach, providing insights into the most effective methods for optimizing material performance and guiding the design of next-generation bio-based polymer nanocomposites. This comparative study aims to enhance the understanding of machine learning applications in materials science and promote the development of sustainable and high-performance composite materials.

Keyphrases: Bio-based Fillers, machine learning, polymer nanocomposites

@booklet{EasyChair:14516,
  author    = {Abey Litty},
  title     = {A Comparative Study of Machine Learning Techniques for Predicting the Performance of Polymer Nanocomposites Incorporating Bio-Based Fillers},
  howpublished = {EasyChair Preprint 14516},
  year      = {EasyChair, 2024}}