This repo contains data and codes to our work on modeling XAI approach for Optimizing Ozone Delignification of Lignocellulose.

This repository contains data and codes to our work on modeling XAI approach for Optimizing Ozone Delignification of Lignocellulose.
Authors:
Muhammad Rizwan a (Email: [e-mail])
Muhammad Ahmad Khan a (Email: [e-mail])
Khurram Shahzad Baig b, * (Email: [e-mail])
a. School of Computing Sciences, Pak-Austria Fachhochschule: Institute of Applied Science and Technology, Mang, Haripur 22621, Pakistan
b. Department of Chemical and Energy Engineering, Pak-Austria Fachhochschule: Institute of Applied Science and Technology, Mang, Haripur 22621, Pakistan.
Corresponding author:Dr. Khurram Shahzad Baig
Department of Chemical and Energy Engineering, Pak-Austria Fachhochschule: Institute of Applied Science and Technology, Mang, Haripur 22621, Pakistan.
Contact: [phone]
Email: [e-mail]
Project description:
This study explores the use of machine learning to enhance ozonation-based lignin removal from lignocellulosic biomass, a key step in biofuel production. Lignin, which makes up about one-third of biomass, hinders reactions with cellulose and hemicellulose. Using Pycaret, 19 machine learning models were tested, with the Extra Trees Regressor providing the best predictions. SHAP analysis was applied to interpret the model results. The findings highlight the potential of machine learning to optimize the delignification process, opening avenues for more efficient and eco-friendly methods in biofuel and chemical production.