An Example Application of Artificial Neural Networks in Archaeology

Explore an innovative application of artificial neural networks in archaeology through this 55-minute lecture from the Santa Fe Institute. Discover how Kelsey M. Reese from the University of Notre Dame uses unsupervised machine learning algorithms to predict occupation periods of undated archaeological sites in the central Mesa Verde region. Learn about the potential of convolutional neural networks in classifying ceramics, categorizing human remains, and identifying artifacts in remotely collected imagery. Gain insights into a novel multi-label classification artificial neural network that quantifies relationships between ceramic assemblages and known occupation dates, resulting in a detailed demographic reconstruction from AD 450-1300. Examine the human-scale experience on the Mesa Verde North Escarpment from AD 890-1300 and understand how this approach offers finer temporal resolution in archaeological research.