Tab2KG: Semantic table interpretation with lightweight semantic profiles

Tabular data plays an essential role in many analytics and machine learning tasks. Typically, tabular does not possess any machine-readable semantics. In this context, semantic table interpretation is crucial for making workflows more robust explainable. This article proposes Tab2KG – a novel method that targets at the of tables with previously unseen automatically infers their semantics to transform them into graphs. We introduce original lightweight profiles enrich domain ontology’s concepts relations represent characteristics. propose one-shot approach relies on these map dataset containing instances ontology. contrast existing approaches, only require instance lookup. property makes particularly suitable which typically contain new instances. Our experimental evaluation several real-world datasets from different application domains demonstrates outperforms state-of-the-art baselines.