ARIADNE: Final Report on Data Mining

Recent years have witnessed a growing interest from archaeological communities in Linked Data. ARIADNE, the Advanced Research Infrastructure for Archaeological Data set Networking in Europe, facilitates a central web portal that provides access to archaeological data from various sources. Parts of these data have been being published as Linked Data, and are currently available in the Linked Open Data cloud. With it, the nature of these data has shifted from unstructured to structured. This presents new opportunities for data mining. While general-purpose software exists, recent studies have revealed the importance of two domain-specific requirements: 1) produce interpretable results, and 2) allow trust in the underlying model. In this work, we investigate to what extend interpretable data mining can contribute to the understanding of linked archaeological data. A case study was held, which involved the mining of semantic association rules over data sets of increasing levels of knowledge
granularity, followed by the qualitative evaluation of these rules by domain experts. Experiments have shown that the approach yielded mostly plausible patterns, some of which were seen as highly relevant.