The trackintel Documentation

Focusing on human mobility data, trackintel provides functionalities for data quality enhancement, integrating data from various sources, performing quantitative analysis and mining tasks, and visualizing the data and/or analysis results. In addition to these core functionalities, packages are provided for user mobility profiling and trajectory-based learning analytics. It is split into the different steps of a typical processing pipeline, and assumes that data is available adhering to the trackintel data model format:

  • Preprocessing (filtering, outlier detection, imputation of missing values)

  • Contextual Augmentation (map matching, trajectory algebra-based context addition)

  • Analysis (extraction of mobility metrics and descriptors, preferences, systematic mobility)

  • Visualization and Communication (generation of maps, charts, etc.)

  • Non-standardized methods and algorithms are explicitly denoted as experimental and (whenever possible) separated from the standardized methods.

For information about the trackintel data models that are used throughout the framework, please refer to the Model page. For a quick deployment to a PostGIS database, you can use the SQL commands given at Data Model (SQL) or run the file found on Github.

Indices and tables