Preventing intersection attacks
AssigneeInternational Business Machines Corporation
InventorsSaritha Arunkumar, Supriyo Chakraborty, Raghu Kiran Ganti, Mudhakar Srivatsa
Life of this patent
Estimating the term…
Abstract
A computer-implemented method (and structure) includes receiving information. The received information is converted into a strictly hierarchical data format. A precision for a releasing the strictly hierarchical data is calculated based on privacy protection levels and a reward for different precision levels. The strictly hierarchical data is sequentially released at the calculated precision.
Claims20
A computer-implemented method for protecting sensitive data against intersection attacks, the method comprising:
- receiving, by a computer, information to be transmitted in a serial-time transmission;
- converting, by the computer, the information into a strictly hierarchical data format;
- calculating a precision for releasing the strictly hierarchical data, based on privacy protection levels and a reward for different precision levels;
- and
- sequentially releasing the strictly hierarchical data at the calculated precision in a time series transmission,
- wherein the calculating of precision uses a Markov Decision Processing (MDP) process implementing a discounted reward model that incorporates the privacy protection levels and which applies a privacy protection level appropriate for a purpose of the serial-time transmissions while maximizing a utility function for the reward based on a granularity of precision in a release, in view of the appropriate privacy protection level, a current number of users, and a probability of changes in the current number of users.
A system, comprising:
- a computer having at least one processor;
- and
- at least one memory device accessible by said at least one processor,
- wherein said memory device stores a set of computer-readable instructions that, when executed by the processor, causes:
- receiving, by the computer, information as a time series;
- converting, by the computer, the information into a strictly hierarchical data format;
- calculating a precision for a release of the strictly hierarchical data, based on privacy protection levels and a reward for different precision levels;
- and
- sequentially releasing the strictly hierarchical data at the calculated precision in a time-series transmission,
- wherein the calculating of precision uses a Markov Decision Processing (MDP) process implementing a discounted reward model that incorporates the privacy protection levels and which maximizes a utility function for the reward based on a granularity of precision in a release, in view of the privacy protection levels, a current number of users, and a probability of changes in the current number of users.
A computer-implemented method for protecting sensitive data against intersection attacks, the method comprising:
- receiving input data from n users as sensitive data to be transmitted in serial time transmissions;
- converting, by a processor on a computer, the information into a strictly hierarchical data format;
- setting parameters of a Markov Decision Processing (MDP) process based on a current number of users providing data to be transmitted, the MDP process implementing a discounted reward model that incorporates the privacy protection levels and which maximizes a utility function for the reward based on a granularity of precision in a release, in view of the privacy protection levels, a current number of users, and a probability of changes in the current number of users,
- exercising the MDP process to calculate a precision for a release of sensitive data;
- and
- providing the sensitive data at the calculated precision to a location based service for transmission of the sensitive data in serial time transmissions in a privacy protection format used by the location based service.
Dependencies are inferred from the claim text.