AI reasoning engine exploiting semantic triples for H-BIM
AI technique applied to H-BIM to add more knowledge to the model by exploiting metadata and media already available on Europeana and external knowledge bases.
Short description
A reasoning engine is a piece of software able to infer logical consequences from a set of asserted facts or axioms. Since 3D H-BIM models are already converted into semantic triples by the service “ICE technologies for converting BIM and H-BIM into semantic data, making them accessible, easier to query and enrich”, this AI technique will be applied to explore if more knowledge can enrich the model. Exploiting metadata and media already available on Europeana and external knowledge, the service will try to retrieve any results that can contribute to the H-BIM model enrichment.
Main features and benefits
Starting from a model uploaded to the INCEPTION platform, users can access external resources available in digital databases and catalogs, such as Europeana, through the integration of the reasoning engine. This engine is designed to infer logical consequences from a set of asserted facts or axioms, enriching the model with additional knowledge.
The Newgrange case study serves as a practical example, showcasing two federated models—a general model and a detailed model—developed within the 5D project by The Discovery Programme and subsequently uploaded to the INCEPTION platform (Figures 1 and 2).
Figure 1: The INCEPTION Platform, A DISCOVERY PROGRAMME project: Newgrange, Meath (IE). General model.
Figure 2: The INCEPTION Platform, A DISCOVERY PROGRAMME project: Newgrange passage, Meath (IE). Detail model.
To explore the metadata of a specific element, users can click on the “metadata” button to access a graphical visualization of the metadata structure (Fig. 3).
Figure 3: The INCEPTION Platform, A DISCOVERY PROGRAMME project: Newgrange passage, metadata graphical navigation.
Within the metadata, a category titled Europeana displays a series of hyperlinks generated by the reasoning engine integrated into the platform.
The reasoning engine processes the available information and the structure of the selected element and its corresponding model. By cross-referencing extracted keywords, it filters data from the Europeana platform (Fig. 4). The result is an automatically generated list of Europeana resources related to the selected element, presented as metadata hyperlinks (Fig. 5).
Figure 4: A bunch of the Newgrange archaeological site resources on the EUROPEANA Platform.
Users can then select any of the provided hyperlinks to be redirected to the corresponding resource page on Europeana, facilitating seamless access to enriched knowledge and external content (Fig. 5).
Figure 5: The INCEPTION Platform, A DISCOVERY PROGRAMME project: Newgrange passage, metadata graphical navigation, detail on the EUROPEANA hyperlinks offered by the INCEPTION Platform reasoning engine.
This workflow demonstrates how the reasoning engine enhances H-BIM models by connecting them with external semantic data and resources, contributing to a more comprehensive and enriched understanding of cultural heritage assets.
