Modelling the past. Logics, semantics and applications of neural computing in archaeology

Abstract

The study of complex archaeological systems through the new Artificial Intelligence and Natural and Neural Computing is a research project which evaluates the historical meaning of the relationships between records of the past as an essentially human construction. It repeats a strong position of Analytical Archaeology, but updates it on the basis of the progress which neurosciences and physics have made in simulating the principles which regulate memory, orientation, classification and mapping of reality. Modelling and simulating the contexts of the past in integrated, parallel, distributed processing through machine learning methods, must make use of a precise encoding of the documents. It takes on an important role in empirical research only when the results produced become the hyper-surface of a network membrane to continue, update, refine or open the analysis itself. After some 30 years of such theoretical, analytical and experimental research, logics, semantics and applications of neural computing maintain their distinct value as a new theoretical approach for the study of dynamic and systemic cultural complexity. They provide a new analytical paradigm for computational modelling in archaeology and an advanced computational method for pattern recognition in archaeometry.