Conductive polymer composites for EMi Shielding Application
Description
This dataset accompanies the manuscript "Explainable Machine Learning-Assisted Inverse Design of Conductive Filler-Reinforced Polymer Composites for Electromagnetic Interference Shielding." It contains a curated literature-derived dataset of Conductive Filler based polymer composite samples.
Files
Steps to reproduce
The dataset was compiled from peer-reviewed research articles reporting the electromagnetic interference (EMI) shielding performance of graphene-based polymer composites. Experimental data were manually extracted from published literature and curated to ensure consistency and completeness. Duplicate, incomplete, and inconsistent records were removed, and the remaining data were standardized into a unified format. The final dataset comprises 456 experimental observations with 11 input descriptors, including polymer matrix, conductive fillers, filler loading, thickness, number of layers, structure, porosity, and annealing conditions, while shielding effectiveness (SE) was used as the target variable. The accompanying Jupyter notebooks (PIML.ipynb and Inverse_Design_Graphene.ipynb) reproduce the preprocessing, machine learning model development, explainable AI analysis, and inverse design workflow described in the associated manuscript. Researchers can reproduce the results by using the provided dataset together with the supplied notebooks.
Institutions
- Pakistan Institute of Engineering and Applied SciencesIslamabad, Islamabad