Configuring privacy policies by formulating questions and evaluating responses
AssigneeInternational Business Machines Corporation
InventorsMarco Pistoia, Omer Tripp
Life of this patent
Estimating the term…
Abstract
Configuring a privacy policy by identifying one or more applications that are installed on a mobile device. A set of privacy-related questions are formulated based upon the one or more identified applications. A response to the set of privacy-related questions is received from a user. One or more classified privacy instances are determined for the user from the received response. A latent, implicit privacy policy is established from the one or more classified privacy instances.
Claims12
A method for configuring a privacy policy, the method comprising:
- identifying one or more applications that are installed on a user's mobile device;
- formulating a set of privacy-related questions based upon the one or more identified applications;
- receiving a response from the user to the set of privacy-related questions;
- determining one or more classified privacy instances for the user from the received response;
- and
- establishing a latent, implicit privacy policy from the one or more classified privacy instances using a machine learning classification procedure;
- receiving a new privacy instance for classification;
- and
- performing a calculation to determine a proximity of a new privacy instance to each of the one or more classified privacy instances for the user, the calculation being performed using a natural distance metric that considers an intensity level of the new privacy instance with respect to a normalized average intensity level of data that the user is willing to release based upon the received response.
A computer program product comprising a non-transient computer-readable storage medium having a computer-readable program stored therein, wherein the computer-readable program, when executed, causes a computer to perform a method for configuring a privacy policy, the method comprising:
- identifying one or more applications that are installed on a user's mobile device;
- formulating a set of privacy-related questions based upon the one or more identified applications;
- receiving a response from the user to the set of privacy-related questions;
- determining one or more classified privacy instances for the user from the received response;
- and
- establishing a latent, implicit privacy policy from the one or more classified privacy instances using a machine learning classification procedure;
- receiving a new privacy instance for classification;
- and
- performing a calculation to determine a proximity of a new privacy instance to each of the one or more classified privacy instances for the user, the calculation being performed using a natural distance metric that considers an intensity level of the new privacy instance with respect to a normalized average intensity level of data that the user is willing to release based upon the received response.
An apparatus for configuring a privacy policy, the apparatus comprising a processor and a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to:
- identify one or more applications that are installed on a user's mobile device;
- formulate a set of privacy-related questions based upon the one or more identified applications;
- receive a response from the user to the set of privacy-related questions;
- determine one or more classified privacy instances for the user from the received response;
- and
- establish a latent, implicit privacy policy from the one or more classified privacy instances using a machine learning classification procedure;
- receive a new privacy instance for classification;
- and
- perform a calculation to determine a proximity of a new privacy instance to each of the one or more classified privacy instances for the user, the calculation being performed using a natural distance metric that considers an intensity level of the new privacy instance with respect to a normalized average intensity level of data that the user is willing to release based upon the received response.
Dependencies are inferred from the claim text.