A framework build for the easy development and research of fake-trajectories and their detection. The framework offers
- 402 different test-trajectories
- 3 algorithms for the creation of random or more elaborate spoof-trajectories
- 4 detection algorithms, based on spatial and time-features
- nodeJs >= v.0.10.0: nodeJs.org
- mongoDB: MongoDb.org
- clone the project if you haven't already: GitHub Repository
- In the root directory run
npm ito install all necessary packages listed in package.json - To install all necessary frontend-packages, run
bower install(you might have to install bower globally first by runningnpm i -g bower) - run GeoVal by either running
npm startto run the current version normally orgrunt serveto run a dev version. For more information on the build system and the basic project-structure visit https://github.com/DaftMonk/generator-angular-fullstack (this project was build using the yeoman angular-fullstack generator)
#Usage instructions
- When everything runs fine, you should see a browser window opening with the GeoVal project
- Log in with the default credentials ([email protected] with password 'admin')
- Go to the 'Admin' page and import the training-data trajectories
- Generate spoof-trajectories of any level (you might have to restart the server for proper initialization of the lvl-3 generator)
- view the trajectories on the map
- automatically analyse all trajectories with the automatic analysis.
- the detection-algorithms can be found in "./server/components/analytics/algorithms"
- the spoof generation algorithms can be found in "./server/components/SpoofFactory/generators"