Computationally Efficient IFC File Comparison Using RDF Graph Paths

Research output: Contribution to conferencePaperScientificpeer-review

Abstract

We study computationally efficient methods calculating changes in IFC models that have been converted to RDF graphs. It is shown how to calculate difference sets that are small enough to be descriptive and usable for synchronization while still using limited computing resources. We present an algorithm that utilizes the neighborhood of an anonymous node to find a probable matching node pair in the second version of the model. By limiting the neighborhood only to the most informative paths in the RDF graph, a balance between the certainty of the matching pair and the size of the neighborhood can be found. The algorithm is compared with commonly available RDF diff tools: Jena, and RDFContextTools RDFC14Ner. When compared, our algorithm provided significantly better results.
Original languageEnglish
Pages599-608
Number of pages10
Publication statusPublished - 2015
EventCIB W78 Conference - Eindhoven University of Technology, Eindhoven, Netherlands
Duration: 27 Oct 201529 Oct 2015
Conference number: 32
http://cib-w78-2015.bwk.tue.nl/

Conference

ConferenceCIB W78 Conference
Abbreviated titleCIB W78
CountryNetherlands
CityEindhoven
Period27/10/201529/10/2015
Internet address

Fingerprint Dive into the research topics of 'Computationally Efficient IFC File Comparison Using RDF Graph Paths'. Together they form a unique fingerprint.

Cite this