Data-Driven Insights in Single-Molecule Break Junction Studies: A Comprehensive Review of the Data Analysis Methods

Single-molecule electronics has emerged as a transformative field at the intersection of chemistry, physics, and nanotechnology, enabling the direct probing of charge transport phenomena at the molecular scale. The break junction technique, which measures conductance across metal-molecule-metal junctions, has become a cornerstone for studying single-molecule dynamics and quantum transport. However, interpreting the large-scale unlabeled conductance traces poses significant challenges. While experimental advancements have improved data acquisition, systematic frameworks for analyzing these complex signals remain underdeveloped. Over the past two decades, substantial efforts have been made to understand single molecules from the perspective of conductance data, and this area remains active in research. Notably, existing reviews have either focused on specific experimental techniques or scattered computational approaches, leaving a critical gap in integrating interdisciplinary methodologies for data analysis. This review primarily focuses on data collected from the break junction technique and discusses data analysis methods across four key aspects: statistical analysis, data simulation, noise spectroscopy analysis, and machine learning techniques. We highlight the essential features of each method and summarize representative works within each category. Furthermore, we provide a perspective on the engagement of large language models (LLMs) in single-molecule science research and discuss their potential impacts on future investigations.