IT Security 2
ISEC2
ISEC2
Set of flashcards Details
| Flashcards | 97 |
|---|---|
| Students | 10 |
| Language | Deutsch |
| Category | Marketing |
| Level | University |
| Created / Updated | 07.03.2016 / 03.11.2021 |
| Weblink |
https://card2brain.ch/box/global_supply_chain_management
|
| Embed |
<iframe src="https://card2brain.ch/box/global_supply_chain_management/embed" width="780" height="150" scrolling="no" frameborder="0"></iframe>
|
7. Privacy - Anonymization
Research demonstrated how little information is required to de-anonymize information (in the Netflix dataset)
Netflix published movie rankings by customers. Remove personal details but it was easy to de-anonymize
7. Privacy - Safeguards
In the field of analytics, customer data can be used responsibly in two ways:
- the appropriate customer permissions (consent) must be in place
- it must be anonymized so that no individual can be identified.
7. Privacy - Anonymization of personal Data
Macro data
- On Mondays on trajectory X there are 160% more pasengers than on Tuesdays
Micro data
- Data is at the granularity of indiciduals
- -> Re-identification risk
7. Priacy - Definitions of Identifier, Quasi-identifier and Sensitive attributes
Identiefier: An attrivute, which uniquely identifies a person Exampels are identity card numbers, account number, ...
Quasi-identifier: A (minimal) set of attrivutes, which may also exist in other tables, and thus may allow linkages identifiying a person up to a certrain degree of precision. For example sex, age, telephone number.
Sensitive attributes: Those attrivutes, which represent information about a person, which shall not be related with that person. For example income, religion, political attiude, health condition
7. Privacy: k-anonymity
Each release of data must be such that every combination of values of quasi-identifiers can be indistinctly matched to at least k respondent.
Original data --anonymation--> 3 anonymous data
7. Privacy: Generalization
- each attribute is associated with a domain to indicate the set of values that the attrivute can assume -> ground domains
- a set of (generalized) domains containing values and a mapping between each domain and domain generalization of it
- for instance, ZIP codes can be generalized by dropping, at each generalization step, the least significant digit
7. Privacy: Suppression (Unterdrückung)
- remove data from the table so that they are not released
- mostly applied at the record level
- -> To moderate the generalization process when a limited number of tuples with less than k occurrences would force a great amount of generalization (outliers)
Both generalization and suspression can be applied at different granularity levels:
- Generalization: cell, attriute
- Supression: cell, attribute, record
Seite 29
7. Privacy k-minimal Generalization
siehe seite 31-33
k-anonymitiy may fail if
- Sensitive values in an equivalence class lack diversity
- The attacker has background knowledge
7. Privacy: Problems Comparision
An increase of the protection of anonymity causes a decrease of the data utility: There is a need for a trade-off between utility and data protection.
7. Privacy: How can we make personal data accessible to third parties in a secure way?
- De-Identification of data does not give a assurance of anonymity
- k-anonymity - protection against identity linkage
- I-diversity - protecion against attrivute disclosure
- Anonymisation methods:
- swapping values, rounding, additive noise,..
- generalization and suppression (outliers)
- Model: k-anonymity
- What are the appropriate parameters for protection and utilty?
PRivacy: Tracking -> Targeting Techniques
- According to browser (Explorer, Firefox, Safari,...) or to provider
- Bandwidth, screen resolution, operatin system
- Geo-/ Regiotargeting (IP adress, GPS,..)
- Referer, cookies,...
- Plug-ins social networks for example Facebook's "like" buttons
7. Privacy: Commercial Fingerprinting
BlueCava, Iovation, ThreatMetrix
- cover all the features described in Panopticlick as well as new features
- catch track browser updates
- fingerprinting capabilities
- browser customization (plugin enumeration,..)
- browser-level user configuration (Cookies enabled,...)
- browser family& version
BlueCava employs multiscreen identification and targeting technologies to provide consumer activity data to large brands.
iovation's fraud prevention software and multifactor authentication solutions help businesses protect against credit card fraud, identity theft, account takeover and other abuses.
7. Privacy: Techniques for locatlon determination
- Access points of the Internet Service Provider
- WLAN
- Mobile radio
- GPS
- Credit cards
-> Location-bases Services require the transmission of user location! Knowledge of detailed location data can expose peoples personal life
-> 4 süatio temporal points are sufficient to identify 95% of the individuals
7. Privacy: Mobile Phones regardet as radiolocation bugs
German politician Malte Spitz published his telecom data from Aug 2009- Feb 2010:
- movements had been captured to 78%
- Zeit Online enriched this data set with public and for everybody easily available data records
7. Privacy: Digital Signage Networks
- collect & analyze detailed information about consumers, their behaviours, and their characteristics
- simple people-counting sensors
- sophisticated facial recognition camers
- to create highly targeted advertisements
But what about notice of collection and informed consent?
7. Privacy_ Escape from Face Recognition
- modifying face -> painting
- jamming photo taking (Störung)
- Privacy visor - glasses that transmit near-infrared signals
- transforming pictures
7. Privacy Zusammenfassung
- There are many different ways to track goods and people
- Facial recognition technology is on the upfront
- We are on the transition form goods to people tracking
- Fantastic new location-base services are emerging but the old challanges remain:
- notice, chois& informed consent
- What is the later use of the collected data?
- link with other data
