Lecture 22 Flashcards
(11 cards)
What is a Sensitive Attribute?
Information People do not want to reveal (e.g. marks on exam)
What is a non-Sensitive Attribute
One that isn’t Sensitive ….. NO SHIT
What is a Quasi-identifier?
A combination of non-sensitive attributes that can be used to identify an idnividual e.g. {Gender, Age, Zip Code}
What is K anonymity?
when every record in a set is indistinguishable from at least k-1 other records with respect to every set of quasi-identifier attributes.
How to achieve K anonymity?
- Remove (suppress) the quasi identifiers compeletely.
- Make the quasi identifiers less specific
What is L diversity?
When the same quasi identifiers point to different sensitive attributes.
Problems with K anonymity?
- Leak information due to quasi identifiers pointing to same sensitive attributes
- Leak information due to background knowledge about the domain.
Benefits of K anonymity?
Worst case can only narrow down to a group of k individuals
Benefits of Location Sharing?
- Precise, tailored location servicies
- Monitoring daily activity for fitness
- Tracking children
- Traffic Monitoring and Navigation
what are the privacy concerns with Location Sharing?
- Personal Safety e.g. Stalking, assault
- Location based profiling e.g. Facebook
- Intrusive interference e.g. individuals personal perferences, health condition etc
What is the trade off between privacy and utility?
Quality of service delivered and analysed data is difficult to maintain the higher the privacy.
