Back

Statistics and Data Science

This research theme focuses on the use of statistical reasoning and data science methods to analyse, interpret, and model archaeological datasets. I am particularly interested in approaches that prioritise clear inference, transparency, and reproducibility, especially when working with heterogeneous, incomplete, or legacy data.

My work in this area spans exploratory data analysis, statistical modelling, and applied machine learning, with an emphasis on understanding uncertainty, assumptions, and limitations in archaeological data. Rather than treating models as black boxes, I focus on interpretable methods and workflows that support critical evaluation and reuse by other researchers.

A central aspect of this theme is the development of reproducible computational pipelines. This includes the use of version controlled code, structured data formats, and well documented analysis steps that allow results to be inspected, replicated, and extended. These practices are informed both by academic research and by professional experience working with large scale data in applied contexts.

I am particularly interested in how statistical and data science methods can be integrated into broader archaeological workflows, including geometric morphometrics, computer vision, and digital documentation.

Overall, this theme aims to bridge methodological theory and practical application, supporting archaeological research to be analytically robust, sustainable, transparent, and accessible to wider research communities.