Machine Learning: A Novel Tool for Archaeology

The nature of archaeology is complex, and is not confined to the humanistic world, but can easily spread to scientific one. Nowadays, it is not uncommonto find archaeologists using new technologies in order to gather, analyze, and interpret data and disseminate the results. It is clear that every aspect ofthe work of the archaeologists is now actively revolutionized by modern technologies. In this framework, the use of machine learning techniquesproves to be extremely interesting. New opportunities for the classification of artifacts, the prediction of site location, and the standardization ofremains analysis can now be offered to archaeology. Although the introduction of machine learning in archaeology dates back to the 1970s,an extensive use in day-to-day work is still far away. Probably, the greatest obstacle is the need for a trained archaeologist expert in machine learning.While the improvements that machine learning could introduce in archeology are clear, the role of the archaeologist in machine learningapplications is not. This limitation should be removed increasing the knowledge in the archaeological community. This work means to offer ananswer to this need providing descriptions of the main algorithms as well as their applications in the archaeological community.