Archaeological and experimental lithic microwear classification through 2D textural analysis and machine learning

Abstract

The paper focuses on introducing 2D texture analysis as a quantitative method for functional analysis in archaeology. Indeed, for the first time, different techniques of quantitative feature extraction and machine learning algorithms applied to the functional analysis of archaeological lithic tools are described and compared. The method presented relies on five techniques of quantitative feature extraction from photographic images and six classification techniques through machine learning algorithms. After creating a training dataset with experimental traces, machine learning models were used to classify 23 images (10 experimental and 13 archaeological). The best result achieved a classification accuracy of 87%, demonstrating the ability to interpret use-wear traces correctly on both experimental and archaeological artefacts regardless of the geological origin of the flint (Sicily in Italy and Sachsen-Anhalt in Germany). The paper proposes to use the method as a fundamental tool in functional analysis to remove subjectivity criteria from traditional analysis and to address issues related to the credibility of the discipline, calibration, standardisation, and reproducibility of methods and results.