This material serves as supplement to "My home is my secret: concealing sensitive locations by context-aware trajectory truncation".
The material contains two folders; each contains an implementation of the algorithm, one in Java and one in Python.
The Python script includes a function (example()) demonstrating how the mechanism class ('STT') may be used. Please note that the clustering algorithm, which was used to generate the protection sets for our experiments, is only available in Java.
The second folder contains a Java implementation of the algorithm ("TrajectoryTruncation") and the clustering algorithm that generates the protection sets required for the truncation algorithm ("GeometricClustering").
The "input"-folder contains the synthetic trajectories that we used for testing the algorithm. It also contains a collection of points ("centroids-utm.shp"). These points serve as input data for the clustering algorithm. We generated them by calculating the centroids of buildings published on OpenStreetMap (OSM).
To run the algorithm, please make sure to include the additional libraries in the "lib" folder in the project's build path.
First, run the Main application in "GeometricClustering", which will generate additonal input data for the truncation algorithm. Afterwards, the S-TT algorithm can be executed by running the "TrajectoryTruncation" application.