Integrating Human Patterns of Qualitative Coding with Machine Learning: A Pilot Study Involving Technology-Induced...

Area of Science:

• Human-computer interaction
• Machine learning applications
• Qualitative data analysis

Background:

• Technology-induced errors pose significant risks in various domains.
• Analyzing incident reports is crucial for understanding error causation.
• Existing methods may not fully capture the nuances of human factors in errors.

Purpose of the Study:

• To develop a reproducible method integrating human qualitative coding with machine learning.
• To apply this method to analyze incident reports.
• To identify factors and types of technology-induced errors.

Main Methods:

• Qualitative coding of incident reports based on technology-induced error and safety literature.
• Application of machine learning algorithms to the coded data.
• Integration of human coding patterns with machine learning analysis.

Main Results:

• Successful application of the integrated method to incident report analysis.
• Identification of key factors contributing to technology-induced errors.
• Accurate categorization of error types.

Conclusions:

• The developed method is reproducible and effective for analyzing incident reports.
• This approach enhances the understanding of technology-induced errors.
• The findings can inform safety improvements and error prevention strategies.