Data processing method and device for realizing privacy protection

Abstract

The embodiment of the invention provides a data processing method for realizing privacy protection, which comprises the following steps of: firstly obtaining a plurality of pieces of original sensor data collected by a plurality of terminal sensors to form an original data record, and obtaining a corresponding service label for a user; then, judging whether the original data record contains sensitive sensor data for reflecting a specific privacy behavior of the user or not; if yes, performing corresponding local replacement by utilizing neutral data to obtain a desensitized data record, and inputting the original data record into a data desensitization model to obtain a desensitization processing result; further, on one hand, based on the desensitization processing result and the desensitized data record, determining desensitization processing loss, and on the other hand, inputting the desensitization processing result into a pre-trained service prediction model to obtain a service prediction result, and determining service prediction loss according to the service prediction result and the service label; and training the data desensitization model based on the desensitization processing loss and the service prediction loss.

Description

technical field [0001] One or more embodiments of this specification relate to the technical field of data security, and in particular, to a data processing method and apparatus for realizing privacy protection. Background technique [0002] At present, many sensors are embedded in personal smart devices, including accelerometers, gyroscopes and magnetometers, etc. The application APP installed on the smart device can access the raw data collected by these sensors, and process the raw sensor data to Realize the interaction with the user, or calculate the indicator data displayed to the user to help the user understand their own activities or physical status. [0003] However, the rich raw sensor data also brings the risk of leaking user privacy. For example, users want to protect the privacy of certain parties, such as gender, age, certain personal behaviors (such as smoking), etc., but some APPs may Illegal or malicious use and speculation of raw sensor data, resulting in ...

AI Technical Summary

Problems solved by technology

[0003] However, the abundance of raw sensor data also brings the risk of leaking user privacy. For example, users want to protect their privacy, such as gender, age, and certain personal behaviors (such as smoking), but some apps may Illegal or malicious use and speculation of raw sensor data, resulting in leakage of user privacy

Images