Aliro: an automated machine learning tool leveraging large language models.

Area of Science:

• Biomedical informatics
• Computational biology
• Data science

Background:

• Biomedical and healthcare fields generate large, complex datasets.
• Analyzing this data with machine learning (ML) is challenging for non-computer scientists.

Purpose of the Study:

• To introduce Aliro, an open-source software package designed to simplify ML analysis.
• To enable researchers to interact with data and retrieve/execute code using large language models (LLMs).

Main Methods:

• Aliro provides a web interface for automated ML analysis.
• It integrates LLMs for code generation and data interaction.
• A pre-trained ML recommendation system assists in algorithm and hyperparameter selection.

Main Results:

• Aliro accelerates the discovery of new data insights.
• It offers visualization tools for evaluated models and data.
• The software automates ML algorithm and hyperparameter selection.

Conclusions:

• Aliro lowers the barrier for ML adoption in biomedical research.
• It empowers researchers to leverage complex data for discovery.
• The open-source package facilitates reproducible and efficient data analysis.