AI development for archaeology, anthropology, and scientific modeling
The AI Development area brings together computer vision, machine learning, model building, and reproducible computational workflows for archaeological and anthropological research.
This work focuses on using AI as a scientific tool rather than as a detached black box. The emphasis is on explicit inference, transparent pipelines, and models that can be evaluated against archaeological, biological, and historical evidence.
Main themes. Computer vision, machine learning, scientific modeling, image-based classification, reproducible tools, and AI literacy for research.
What this area is doing
- Developing computational tools that support classification, measurement, and comparison of archaeological materials.
- Linking AI workflows to morphometrics, Bayesian inference, and formal model building instead of treating them as standalone prediction systems.
- Using reproducible code and transparent validation so model behavior can be interpreted and questioned.
- Extending AI work into public-facing questions about scientific reasoning, education, and responsible use.
Representative publications and outputs
- Machine learning, bootstrapping, null models and why we are still not 100% sure which marks were made by crocodiles (2022)
- A New Approach to the Quantitative Analysis of Bone Surface Modifications: the Bowser Road Mastodon and Implications for the Data to Understand Human-Megafauna Interactions in North America (2023)
- geomorph: an R package for the collection and analysis of geometric morphometric shape data (2013)
- AI, education, and public scholarship (TIME author page)
AI links
The AI Development area links computational modeling, computer vision, and scientific inference across multiple QuantLab projects.