Assignment of Regioisomers Using Infrared Spectroscopy: A Python Coding Exercise in Data Processing and Machine Learning

Machine learning is a set of tools
that are increasingly used in
the field of chemistry. The introduction of potential uses of machine
learning to undergraduate chemistry students should help to increase
their comprehension of and interest in machine learning processes
and can help support them in their transition into graduate research
and industrial environments that use such tools. Herein we present
an exercise aimed at introducing machine learning alongside improving
students’ general Python coding abilities. The exercise aims
to identify the regioisomerism of disubstituted benzene systems solely
from infrared spectra, a simple and ubiquitous undergraduate technique.
The exercise culminates in students collecting their own spectra of
compounds with unknown regioisomerism and predicting the results,
allowing them to take ownership of their results and creating a larger
database of information to draw upon for machine learning in the future.

机器学习（Machine Learning）是一类日益广泛应用于化学领域的工具集合。将机器学习的潜在应用场景介绍给化学专业本科生，有助于提升其对机器学习流程的理解与兴趣，同时可辅助他们顺利过渡到使用此类工具的研究生科研与工业工作环境中。为此，我们开发了一项教学实践活动，旨在向学生介绍机器学习知识的同时，提升其通用Python编程能力。该实践活动的目标为：仅通过红外光谱（Infrared Spectra）识别二取代苯体系的区域异构性（regioisomerism），而红外光谱是本科生实验中一项简单且常用的测试技术。本次实践活动的收尾环节为：学生自行采集具有未知区域异构性的化合物的光谱并完成结果预测，这既能让学生自主掌握自身的实验结果，也能为未来开展机器学习相关研究积累可供参考的大规模信息数据库。