Large Language Models (such as ChatGPT) as Tools for Machine Learning-Based Data Insights in Analytical Chemistry

Area of Science:

• Analytical Chemistry
• Artificial Intelligence
• Spectroscopy

Background:

• Deep learning, including convolutional neural networks (CNNs), has transformed various scientific fields.
• Natural Language Processing (NLP) has advanced significantly with the development of Large Language Models (LLMs).
• AI techniques are increasingly used to enhance data analysis in analytical chemistry.

Purpose of the Study:

• To demonstrate the application of a Large Language Model (LLM) for multivariate data analysis.
• To showcase the use of an LLM via a smartphone for interactive data analysis.
• To explore the potential of LLMs in processing and analyzing hyper-spectral imaging data from laser-induced breakdown spectroscopy (LIBS).

Main Methods:

• Utilized a Large Language Model (LLM) through a smartphone interface.
• Applied the LLM to a hyper-spectral imaging dataset obtained from laser-induced breakdown spectroscopy (LIBS).
• Leveraged the LLM's capability to interactively analyze data and generate/execute code.

Main Results:

• Demonstrated successful multivariate data analysis of LIBS hyper-spectral imaging data using an LLM.
• Showcased the ability of LLMs to process and analyze complex scientific datasets interactively.
• Confirmed that LLMs can automatically generate and execute code for data analysis tasks.

Conclusions:

• LLMs show significant potential for revolutionizing data analysis in analytical chemistry.
• Smartphone-based LLM analysis offers an accessible and interactive approach to complex scientific data.
• LLMs are anticipated to play an increasingly important role in the future of analytical chemistry research and practice.