Geo-Semantic User Profiling & Privacy

This research area investigates how location data from mobile devices and social networks can be used to build semantically rich profiles of users — and critically, how to protect user privacy when such profiles can reveal sensitive personal information. The work spans from 2011 to 2017 across four interconnected PhD projects.

Research Journey

Research Themes

• Place Affordance Profiling: Extracting user profiles from GPS logs based on what activities places enable
• Geo-Folksonomy Analysis: Discovering place semantics from collaborative tagging
• Spatiotemporal Modelling: Capturing how user preferences evolve over time
• Location Recommendation: Personalised place suggestions based on semantic profiles
• Privacy Threat Modelling: Understanding risks of location disclosure
• Privacy Awareness: Enabling informed consent about location sharing

Research Projects

PhD Students

• Ahmed Al-Azzawi (2011): Place affordance profiling from GPS mobility logs
• Ehab ElGindy (2013): Place semantics from geo-folksonomies
• Soha Mohamed (2017): Spatiotemporal user and place modelling on LBSNs
• Fatma AlRayes (2017): Location privacy awareness and threat modelling

Key Contributions

• Personal Gazetteers and PGML markup language for encoding user profiles
• Place type and activity ontologies derived from social tagging data
• Static and dynamic user models for location-based social networks
• User and place similarity measures based on semantic interactions
• Privacy threat models for geo-social network data
• Privacy feedback solutions enabling informed location disclosure decisions

Data Sources

Research in this area utilised data from:

• GPS mobility traces from personal mobile devices
• Geo-folksonomies (Tagzania and similar platforms)
• Location-Based Social Networks (Foursquare)
• User check-ins, tips, and social interactions