Event Retrieval from Mobile Devices with Location Anonymization Satisfying Geo-Indistinguishability Using Real Population Data

In recent years, several services have enabled information on the vicinity of the current location by providing location information from mobile devices. However, there is a privacy concern because the location information of the user, which is personal information, is transmitted directly to the server and used for information analysis. Therefore, in this study, we use real population data to calculate a dummy user location that satisfies Geo-Indistinguishability, a technique that extends differential privacy to location anonymization and uses it to perform user location anonymization by querying the server. We propose a method for retrieving event information from X (formerly Twitter), a type of social networking service, while protecting user positions and satisfying geographical constraints.