nyrio/geolocation-based-user-profiling

School project: profiling users from geolocation data

Big Brother - profiling users from geolocation data

Introduction and architecture documents

See report.pdf.

Installation

Loading the data

Our tool reads data from an InfluxDB database. You need to setup a database. An easy way to get started is to use Docker Compose with docker-compose.yml. The default settings of our scripts match the default settings of a local InfluxDB database (on localhost:8086 with username root and password root).

The script populate_influxdb.py loads data from a file to the database. You will want to use it with the --gps option (use -h to see a short manual). The file must contain on each line the following tab-separated values: user id, date, longitude, latitude. The data will be added in the gps measurement of the rawdata database. Each data point has a tag id with the user id, and fields lat and lon (and of course a timestamp time).

The extract_sample.py is intended to extract from a data file the points corresponding to the given user and time range. The only purpose of this file is to reduce the size of the data loaded in the database as loading gigabytes of data was quite long. There are also options in our tool to load only a given time range from the db.

Compiling the tool

In the src/core directory, run make cli. This will create a cli executable and cache the compilation files in the build folder.

Using custom parameters

The config.txt file contains parameters read at runtime by our tool. As your data is different from ours, you will probably need to use custom parameters. Check the comments in the file for more information.

Running the program

You can simply run ./cli. If you move the executable, you also need to move the config file. To see the available commands, type help in the tool's command-line interface.

Disclaimer

We cannot be held responsible for any malicious of this software.

This is a school project: this program is distributed without any warranty; without even the implied warranty of merchantability or fitness for a particular purpose. The program might fail if it cannot connect to the database or an external API, or with data that doesn't match the expected format.