Master in Big Data Analytics & Social Mining, University of Pisa.

• Mirco Nanni (ISTI-CNR)
• Luca Pappalardo (ISTI-CNR)

In these hands-on lessons, we present an overview on the fundamental principles underlying the analysis of mobility data, using scikit-mobility, a specific Python library designed by the tutorial presenters.

In particular, we see how to visualize trajectories, flows and tessellations, how to clean raw mobility data by using standard techniques proposed in the mobility data mining literature, and how to analyze mobility data by using the main measures characterizing human mobility patterns (e.g., radius of gyration, daily motifs, mobility entropy).

Lessons

• Lesson 1: Intro and Data Structures
• Lesson 2: Preprocessing mobility data
• Lesson 3: Mobility measures

Practice

• Practice 1
• Practice 2
  • traccia
  • solved
• Practice 3
  • traccia
  • solved
• Read Google Timeline Data

• Repository: https://github.com/scikit-mobility/scikit-mobility

• Documentation: https://scikit-mobility.github.io/scikit-mobility/

• Paper: https://arxiv.org/abs/1907.07062

• Jupyter Notebook: https://jupyter.org/

• To run notebooks in Jupyter using slideshow, install RISE

• Download Geolife Dataset, unzip it, and put it in folder data

• Download San Francisco cabs data, and put the file in the folder data

• Download your Google Location data, and put the json file in folder data

• Download data of taxis in Rome

• How to make folium Custom Icons: https://ocefpaf.github.io/python4oceanographers/blog/2015/11/02/icons/
• Cloropleth maps in folium: https://python-graph-gallery.com/292-choropleth-map-with-folium/
• Adding the geographic scale in a folium map: python-visualization/folium#414