Chemical identification using AI

• Main Author: Sua, Heng Hang
• Other Authors: Cai Yiyu
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2020
• Subjects: Engineering::Aeronautical engineering
• Online Access: https://hdl.handle.net/10356/140421
• Institution: Nanyang Technological University

Machine learning is a method of data analysis and model building that allows machines to learn from data, identify patterns and make decisions with minimal human interventions. In this work, supervised machine learning will be used to train a model from data sets consisting of power absorption spectra of various pure chemicals and chemicals mixtures in different packages obtained from terahertz time-domain spectroscopy, and subsequently predict the chemical or composition of an unknown data. Various models such as Support Vector Machine (SVM) and Random Tree Classifier (RTC) were adopted and the accuracy of identification of pure chemicals and chemical mixtures are 93% and 70% respectively. Proven that using machine learning is feasible to identify chemicals, a prototype that incorporates both hard and soft ware is built, and the process pipeline as well as future consideration are explained and proposed.