User-centred privacy inference detection for smart home devices

Enhancing user awareness on inferences obtained from fitness trackers data

“…In our previous work, we introduced PrivacyEnhAction, a web application that aims to inform the users about potential privacy vulnerabilities that emerge from the use of smart water meters and motion sensors (Kounoudes et al 2021). The frontend of PrivacyEnhAction has been designed following the structure of a web page using html and css styles.…”

“…After the possible inferences that can be extracted from fitness trackers data have been identified, the next step is to find which inferences can be drawn from the data collected from the specific fitness trackers in this study. We also describe the methodology we used in this study in order to collect, examine and analyse the data in the fitness trackers scenarios, following the methodology we proposed in our previous work (Kounoudes et al 2021) adjusted to suit the current study's needs, which can be applied in other IoT scenarios with minor modifications.…”

“…This is the focus of this work, where by concentrating on fitness trackers from three brands, namely Fitbit, Garmin and Xiaomi, we investigate if the analysis and exploitation of the data collected by those trackers can lead to the extraction of inferences about the owners routines, health status or other sensitive information. We utilise a data inference framework introduced in our previous work (Kounoudes et al 2021), where by using a number of machine learning, statistical analysis and modelling techniques we aim to verify that such inferences are possible in order to raise user awareness about them. These techniques are applied in the PrivacyEnhAction privacy tool introduced in our previous work, a web application through which the users can analyse data collected from their smart devices (smart water meters or motion sensors) with the objective to be informed about potential privacy vulnerabilities and possible inferences that emerge from the use of these devices, and thereupon to be able to change and set their user privacy preferences on their devices appropriately, contributing in this way to the personalisation of the provided services, in connection with their personal data.…”

We Need a “Building Inspector for IoT” When Smart Homes Are Sold

“…For example, Kounoudes et al showed that surprisingly detailed home-occupant behavior can be identified using seemingly unimportant data from a simple IoT water flowmeter. 10 It might not be readily apparent to a nonexpert user that logs from the water flowmeter should be erased. A comprehensive inventory of devices left behind, however, can give an expert home IoT inspector clues about what should be addressed.…”