Privacy Loop-Holes of Slicing: Privacy Preservation Of High-Dimensional Data

Mona Tanwar, Jitendra Kumar


In this information age, person specific data is shared on a large scale by individuals on various sites which is collected by various government agencies and private companies for knowledge based decision making, research, analysis and other specific purposes. Such data contains personal information of individuals which should be preserved from disclosure as it is a risk to individual’s privacy if it gets into wrong hands. There have been incidents where adversaries got access to personal information of individuals and they misused it. Since a decade, data has increased at an explosive rate because of much increased interaction of users worldwide with internet where they keep sharing their sensitive information on various shopping sites, online billing sites, online social networking sites, search engines, etc. Hence user specific information is shared much more in comparison to before. Privacy has become a big concern while data sharing and publishing. To ensure that the privacy of sensitive information of users is not violated, certain measures are taken before the data is shared or published. Among them Anonymization, Bucketization and Slicing are the most popular techniques. For data of high dimensions i.e. big data, slicing serves the purpose much better than the rest techniques because of limitations with the other techniques which cannot be ignored. In this paper, we discuss slicing in detail and also highlight its privacy limitations in some particular scenarios when it is applied to high dimensional data.


Slicing, Privacy, Big data, Social Networks, PPDP

Full Text:



  • There are currently no refbacks.