Potential of using Machine learning Technology to predict site location of Burial Sites in the Åland Islands.

Sammanfattning

The most recent advancement in the subdiscipline of Digital Archaeology is the introduction of Artificial Intelligence (AI) in conjunction with Machine Learning (ML) as a tool to predict potential archaeological sites in a landscape. Furthermore, the development of new LiDAR (Light Detection and Ranging) technology with a resolution of 5 laser–points/m2 in Åland (as well as large parts of Finland) brings the possibility of analysing hidden archaeological features hidden beneath thick forests and shrubbery. Åland, being a cross–cultural region between Sweden and Finland with a rich archaeological footprint influenced by both countries, may have forgotten archaeological sites hidden in the thick forests of the mainland or among the thousands of islands scattered in the archipelago. This study seek to explore the possibility of using a semi–automated deep learning approach to predict potential burial mounds and cairns in Åland, utilizing the latest high–resolution LiDAR dataset. Although only 4 of the 14 study areas were thoroughly surveyed, the AI/ML system was able to identify sites interpreted to be potential burial sites. As archaeological sites are provided legal under the Ålandic cultural heritage act called Landskapslagen (1965:9) om Fornminnen, it is important to find any potentially significant archaeological site to ensure its legal protection. Therefore, the study suggest that AI/ML systems have the potential to revolutionize the fields of archaeology and Cultural Heritage Management in Åland, with the hopes that similar studies can explore the possibility of finding forgotten archaeological sites in the archipelago region where little have been previously surveyed.