Key points are not available for this paper at this time.
In a modern smart building, many aspects of the use can be monitored using sensing technologies. This enables a high number of data-driven applications used for many tasks, such as indoor comfort, energy efficiency, and space utilization. Open data sharing enables more robust data-driven applications for optimizing building operations. To enable such data sharing effort, there is a need for performing a privacy risk assessment for analyzing the inherent potential ethical and privacy risks that can be posed for occupants and the organization operating in the building. It is increasingly difficult to identify the inference capabilities of modern machine learning methods e.g. for estimating occupancy from CO2 datasets. In this paper, we design and implement an open source ontology-based tool-chain that can be used as part of the privacy assessment to identify potential privacy risks. This tool-chain takes in a model of the dataset that is being considered for sharing and creates a privacy risk report. We evaluate the tool-chain using five real-world datasets and compares the analysis with the data custodian. The results obtained show that the tool-chain can identify more risks, than a human data curator, and thus, there is a need for such a tool to support privacy risk analysis.
Schwee et al. (Wed,) studied this question.