Page 48 - Read Online

Li et al. J. Mater. Inf. 2025, 5, 29 https://dx.doi.org/10.20517/jmi.2024.103 Page 7 of 18

Table 2. The symbolic definitions of variables in aging experimental data
Shore
IR 2,914 IR 2,840 IR 964 IR 1,736 IR 2,914 IR 2,840 IR 964 IR 1,736 XLD Antioxidant Plasticizer Elongation at
Property -1 -1 -1 -1 -1 -1 -1 -1 hardness
cm (S) cm (S) cm (S) cm (S) cm (C) cm (C) cm (C) cm (C) (mol/L) content (mg/g) content (mg/g) (HA) fracture (%)
Variable ω 1 ω 2 ω 3 ω 4 v 1 v 2 v 3 v 4 XLD LC a LC p Hardness Fe
name

IR: Infrared spectroscopy, S: surface, C: cross-section; XLD: crosslink density.

LF-NMR spectroscopy is used to directly quantify the XLD, a crucial parameter that reflects the degree of polymer crosslinking within the material, influencing
its mechanical properties.

Liquid chromatography is employed to directly measure the content of antioxidants and plasticizers in the sample. The presence of antioxidants (hindered
phenol-type compounds) helps to mitigate oxidative degradation, while plasticizers (fatty esters and phthalate esters) are crucial in modifying the flexibility
and hardness of the polymer.

Regarding mechanical properties, Fe is defined as the ratio of the material’s elongation at the point of fracture to its original length, typically expressed as a
percentage. Shore hardness is a measure of the material’s surface hardness, determined using a Shore hardness tester. These mechanical properties provide
valuable insights into the material’s performance under different aging conditions and are key to understanding the relationship between molecular structure
and macroscopic behavior. The distribution of all aging variables is shown in Supplementary Figure 1.

Baseline regression methods
Multiple linear regression
Multiple linear regression (MLR) posits the existence of a linear relationship between independent variables (features) and a dependent variable (target).
Typically, it is represented as a linear combination of features plus an intercept term, optimized by minimizing a loss function - such as the sum of squared
residuals.

Support vector machine regression
Support vector machine regression (SVR) extends the principles of support vector machines to handle continuous target variables. SVR aims to identify a
function that deviates from the true outputs by no more than a pre-specified tolerance, while minimizing the overall model complexity. The framework relies
on kernel functions to efficiently capture nonlinear relationships by mapping inputs into a higher-dimensional space.