Directions in Big Data Anonymisation

30 mins 20 secs, 441.35 MB, MPEG-4 Video 640x360, 29.97 fps, 44100 Hz, 1.93 Mbits/sec

Share this media item:

Embed this media item:

<iframe width="_width_" height="_height_" src="https://sms.cam.ac.uk/media/2376293/embed" frameborder="0" scrolling="no" allowfullscreen></iframe>

Choose size:

About this item

Available Formats

About this item

Description:	Domingo-Ferrer, J (Universitat Rovira i Virgili) Monday 5th December 2016 - 14:40 to 15:10


Created:	2016-12-13 12:51
Collection:	Newton Gateway to Mathematics
Publisher:	Isaac Newton Institute
Copyright:	Domingo-Ferrer, J
Language:	eng (English)


Abstract:	The explosion of big data opens such huge analytical and inferential possibilities that they may allow modeling the world and predicting its evolution with great accuracy. The dark side of such data abundance is that it complicates the preservation of individual privacy: big data largely feed on the digital trace of our activities. Facing the tension between big data and privacy, we find two extreme positions that strive for hegemony: on the one side, the nihilists claim that it is delirious to try to maintain one's privacy in the big data world, and that the best we can hope for is that our data are not misused (if this means anything); on the other hand, the fundamentalists propose privacy protection methods so drastic that their application would destroy nearly all the analytical interest of big data. We will survey these extreme positions and we will describe a midway path, which we believe more balanced and desirable. This path is based on identifying the utility and privacy requirements of big data and trying to satisfy them through an evolution of the statistical disclosure control methods developed in the last 40 years. We will also briefly touch on transparent, local and collaborative anonymization as ways to reduce the power of the data controller in front of individual subjects.

Abstract:

The explosion of big data opens such huge analytical and inferential possibilities that they may allow modeling the world and predicting its evolution with great accuracy. The dark side of such data abundance is that it complicates the preservation of individual privacy: big data largely feed on the digital trace of our activities. Facing the tension between big data and privacy, we find two extreme positions that strive for hegemony: on the one side, the nihilists claim that it is delirious to try to maintain one's privacy in the big data world, and that the best we can hope for is that our data are not misused (if this means anything); on the other hand, the fundamentalists propose privacy protection methods so drastic that their application would destroy nearly all the analytical interest of big data. We will survey these extreme positions and we will describe a midway path, which we believe more balanced and desirable. This path is based on identifying the utility and privacy requirements of big data and trying to satisfy them through an evolution of the statistical disclosure control methods developed in the last 40 years. We will also briefly touch on transparent, local and collaborative anonymization as ways to reduce the power of the data controller in front of individual subjects.

Available Formats

Format	Quality	Bitrate	Size
MPEG-4 Video *	640x360	1.93 Mbits/sec	441.35 MB	View	Download
WebM	640x360	879.45 kbits/sec	195.39 MB	View	Download
iPod Video	480x270	522.17 kbits/sec	116.01 MB	View	Download
MP3	44100 Hz	249.83 kbits/sec	55.54 MB	Listen	Download
Auto	(Allows browser to choose a format it supports)