University of Twente Student Theses

Login

Addressing the limited adoption of Semantic Web data integration in AEC industry : Exploration of an RDF-based approach depending on available ontologies and literals matching

Patsias, D. (2019) Addressing the limited adoption of Semantic Web data integration in AEC industry : Exploration of an RDF-based approach depending on available ontologies and literals matching.

[img] PDF
1MB
Abstract:The limitations of Building Information Modeling (BIM) regarding the complete and integrated management support for building data, which are vast, diverse and distributed across different sources, systems and actors and which often need to be combined for the purpose of several analyses, stimulated the use of Semantic Web. The latter enables the uniform representation of heterogeneous data in structured graphs, using the Resource Description Framework (RDF) and common languages that describe relationships among these, namely ontologies. By deploying Semantic Web technologies, several research streams have focused on the integration of BIM with cross-domain data such as GIS, product and material data, sensor and cost data among others, by suggesting the development of several ontologies that represent concepts and relationships from different domains and the establishment of semantic links between them. At the same time, people without prior relevant awareness, perceive the concept of ontologies and tasks related to them as something too complicated. This is considered the main reason for the slow adoption of Semantic Web technologies, which is obvious within the AEC-FM industry as well. In this paper, the feasibility of a data integration approach that uses available ontologies and avoids ontology alignment is explored. For that reason, an approach based on the RDF representation of diverse datasets, the semantic description of these using available ontologies and the integration of these by matching literals between datasets, instead of establishing semantic links, is presented. A conceptual framework is developed and tested in the context of a simple but typical scenario where BIM has to be integrated with heterogeneous data sources in order to perform several analyses. The results of the evaluation showed that although proven successful for linking datasets, the unlikelihood of the existence of ontologies that fully describe the datasets and cover the data needs of every organization, the requirements regarding the lexical and data type similarity of the literals, the required prior specific knowledge about the data and the resulting complexity of the queries, render this approach hard to be properly applied and generalized. The discussion and conclusion sections, elaborate on the tradeoffs of the approach and the lessons learned regarding the requirements, the potential and the challenges for engineering companies in their effort to comprehend and successfully deploy Semantic Web technologies within their everyday data management practices.
Item Type:Essay (Master)
Faculty:ET: Engineering Technology
Subject:56 civil engineering
Programme:Construction Management and Engineering MSc (60337)
Link to this item:https://purl.utwente.nl/essays/80019
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page