Interpretable Prediction and Analysis of PVA Hydrogel Mechanical Behavior Using Machine Learning

Abstract

Polyvinyl alcohol (PVA) hydrogels have emerged as versatile materials due to their exceptional biocompatibility and tunable mechanical properties, showing great promise for flexible sensors, smart wound dressings, and tissue engineering applications. However, rational design remains challenging due to complex structure-property relationships involving multiple formulation parameters. This study presents an interpretable machine learning framework for predicting PVA hydrogel tensile strain properties with emphasis on mechanistic understanding, based on a comprehensive dataset of 350 data points collected from a systematic literature review. XGBoost demonstrated superior performance after Optuna-based optimization, achieving R² values of 0.964 for training and 0.801 for testing. SHAP analysis provided unprecedented mechanistic insights, revealing that PVA molecular weight dominates mechanical performance (SHAP importance: 84.94) through chain entanglement and crystallization mechanisms, followed by degree of hydrolysis (72.46) and cross-linking parameters. The interpretability analysis identified optimal parameter ranges and critical feature interactions, elucidating complex non-linear relationships and reinforcement mechanisms. By addressing the "black box" limitation of machine learning, this approach enables rational design strategies and mechanistic understanding for next-generation multifunctional hydrogels.

Keywords: PVA hydrogel; SHAP analysis; feature importance; hyperparameter optimization; interpretable machine learning.

Figure 1

Distribution analysis of continuous input features. (a) PVA_MW, (b) PVA_DH, (c) PVA_Conc, (d) FC, (e) CLM, (f) CC.

Figure 2

Pearson correlation matrix of input features.

Figure 3

Machine learning model performance comparison for tensile strain prediction. (a) AdaBoost, (b) GBR, (c) KNN, (d) SVM, (e) XGBoost, (f) MLP.

Figure 4

(a) Predicted vs. true values with performance metrics, (b) feature importance ranking, (c) residual analysis for training and test datasets.

Figure 5

(a) Global SHAP feature importance values, (b) SHAP summary plot showing individual feature contributions across all samples, (c–e) SHAP dependence plots for PVA_Conc, PVA_MW, and CC, respectively, with color coding indicating interaction effects.