Dogs in Space: An Application of Machine-Learning Geometric Morphometric Analyses for Species Determination of Large Canids Using Mandibles

Summary

This is an abstract from the "Machine-Learning Approaches to Studying Ancient Human-Environmental Interactions" session, at the 90th annual meeting of the Society for American Archaeology.

A persistent issue in zooarchaeology is the differentiation of domesticated dogs from wolves and coyotes from fragmentary archaeological remains. This is particularly problematic in regions where size cannot be used as a factor, such as the North American northern Great Plains. This poster presents the use of ancient DNA, traditional osteometrics, qualitative observations, and geometric morphometrics to create a training sub-set of an assemblage of dogs, wolves, and coyote mandibles of varying completeness. This training dataset is then used to create probabilistic species determination hypotheses for the rest of the assemblage using a K-Nearest Neighbor algorithm and a series of geometric morphometric analyses.

This Resource is Part of the Following Collections

• Society for American Archaeology 90th Annual Meeting, Denver, CO (2025)   •
• Machine-Learning Approaches to Studying Ancient Human-Environmental Interactions

Keywords

Record Identifiers

Abstract Id(s): 51333