The conference paper “Dihedron Algebraic Embeddings for Spatio-temporal Knowledge Graph Completion”, co-authored by Mojtaba Nayyeri, researcher of the Analytic Computing department, proposes an embedding model that uses Dihedron algebra for learning such spatial and temporal aspects. According to the authors: “Many knowledge graphs (KG) contain spatial and temporal information. Most KG embedding models follow triple-based representation and often
neglect the simultaneous consideration of the spatial and temporal aspects. Encoding such higher dimensional knowledge necessitates the consideration of true algebraic and geometric aspects. Hypercomplex algebra provides the foundation of a well-defined mathematical system among which the Dihedron algebra with its rich framework is suitable to handle multidimensional knowledge. The evaluation results show that our model performs significantly better than other adapted models”
Diellza Elshani, a Ph.D. student from the Institute of Computational Design and Construction (ICD) and the Cluster of Excellence Integrative Computational Design and Construction for Architecture (IntCDC) of the University of Stuttgart, presented two papers in the Linked Building Data workshop, which is co-located with the conference. Her papers are a result of the collaboration with the Analytic Computing department.
The first paper, “Towards Better Co-Design with Disciplinary Ontologies: Review and evaluation of data interoperability in the AEC industry” was written under the supervision of professors Thomas Wortmann (from the Institute of Computational Design and Construction) and Steffen Staab (head of the Analytic Computing department). This paper reviews and evaluates the interoperability paradigms in centralized, decentralized, and federated data, and it gives architecture, engineering, and construction application examples for each approach. It discusses data schemas and interoperability tools and relates them to Semantic Web standards. According to Diellza: “Developing modular disciplinary ontologies to support collaborative design tools can provide a solid ground to share and exchange data flexibly.
The second paper, “Knowledge Graphs for Multidisciplinary Co-Design: Introducing RDF to BHoM”, results from the collaboration of Buro Happold (Alessio Lombardi and Al Fisher), Analytic Computing (Steffen Staab and Daniel Hernandez) and Computational Design and Construction (Diellza Elshani and Thomas Wortmann). This paper presents a concrete proposal to address the issues highlighted in the first paper using the open-source platform Buiding Habitat object Model (BHoM). The paper explores and presents methods to recast BHoM's object model as a knowledge graph using Semantic Web standards and presents a BHoM RDF prototype as a potential solution to the interoperability problem of multidisciplinary design in architecture, engineering, and construction. When combined with Semantic Web standards, the BHoM framework can increase the use of knowledge graphs in this industry, improving data interoperability and assisting design decisions through reasoning.
Finally, Cosimo Gregucci, a Ph.D. student that is initiating his studies in Analytic Computing, attended the conference to gain experience as a researcher. According to him, the conference was an opportunity to attend interesting talks and to connect with other researchers.