Investigation of dielectric studies of paracetamol-diethylamine solutions: Experimental and machine learning approach.

This paper discusses about the dielectric studies of binary mixtures of paracetamol (PCM) and Diethylamine (DEA). Parallel resistance (R) and Parallel capacitance (C) measured using a precision LCR meter over a frequency range of 20 Hz-2 MHz at four distinct temperatures, starting from 293.15 K and increasing by 10 K for each subsequent measurement. These experimental parameters were used to compute the complex dielectric function, from which electrical properties like complex conductivity, complex impedance and complex electrical modulus-were derived. In addition to conventional analysis, machine learning (ML) models were implemented to predict dielectric constant (ε') and dielectric loss (ε″) values based experimental inputs, with their predictive performance significantly enhanced through Bayesian hyperparameter optimization. This dual approach of combining experimental data with ML modelling offers a novel methodology for efficient and accurate characterization of dielectric systems. The added value of this study lies in its ability to bridge physical measurements with computational predictions, reducing experimental workloads and improving generalization in similar systems. The findings have potential applications in material science, pharmaceuticals, and electronic device modelling. This study demonstrates that ML assisted dielectric analysis can serve as a powerful tool in predictive material characterization.