783AA Geospatial Analytics A.A. 2022/23

[WARNING]: This course substitutes “Big Data Analytics” from the academic year 2022/23 on.

Instructors:

Luca Pappalardo
- luca [dot] pappalardo [at] isti [dot] cnr [dot] it
- KDD Laboratory, ISTI-CNR, Pisa
- http://www-kdd.isti.cnr.it

Mirco Nanni
- mirco [dot] nanni [at] isti [dot] cnr [dot] it
- KDD Laboratory, ISTI-CNR, Pisa
- http://www-kdd.isti.cnr.it

Tutors:

Giuliano Cornacchia, PhD student, University of Pisa
Giovanni Mauro, PhD student, University of Pisa
Daniele Gambetta, PhD student, University of Pisa

Hours and Rooms

Day of Week	Hour	Room
Thursday	11:00 - 13:00	Room Fib E
Friday	09:00 - 11:00	Room Fib C1

Beginning of lectures: 15 September 2022
End of lectures: 2 December 2022
Possible lessons recovered: 5–16 December 2022

The lectures will be only in presence and will NOT be live-streamed

A Telegram channel will be used to post news and other stuff related to the course:

NEWS

Master theses available on Geospatial Analytics. Contact the professors for further information.
Exam dates: January 27th, 2023 and February 24th, 2023, in aula Faedo at ISTI-CNR , Pisa.
Exam instructions:
- each student will present the project, i.e., submitted notebook(s) (in max. 15 minutes)
- during/after the presentation, we'll ask some questions about the project
- and some questions about the course in general (mainly about the topic related to the project)
- (for the computationally expensive code blocks, please pre-run them and show the outputs only).
List of projects
Projects assignment
Evaluation of homeworks 1.x and 2.x
Evaluation of homeworks 3.x
Evaluation of homeworks 4.x, 6.x, and 8.x
Evaluation of homeworks 5.x and 7.x
Evaluation of homeworks 9.x and 10.x

Learning goals

The analysis of geographic information, such as those describing human movements, is crucial due to its impact on several aspects of our society, such as disease spreading (e.g., the COVID-19 pandemic), urban planning, well-being, pollution, and more. This course will teach the fundamental concepts and techniques underlying the analysis of geographic and mobility data, presenting data sources (e.g., mobile phone records, GPS traces, geotagged social media posts), data preprocessing techniques, statistical patterns, predicting and generative algorithms, and real-world applications (e.g., diffusion of epidemics, socio-demographics, link prediction in social networks). The course will also provide a practical perspective through the use of advanced geoanalytics Python libraries.

The assessment of the course consists of: (1) an oral exam, aimed to test the knowledge acquired by the student during the course; (2) exercises to be done during the course; (3) the development of a project to test the practical ability acquired during the course.

Topics:

Spatial Reference Systems
Data formats
Trajectory and Flows
Spatial Tessellations
Open-source tools for geospatial analysis
Digital spatial and mobility data
Preprocessing mobility data
Privacy issues in mobility data
individual and collective mobility laws
Next-location and flow prediction
Trajectory and flow generation
Applications

Module 1: Spatial and Mobility Data

Fundamentals of Geographical Information Systems
- Geographic coordinates systems
- Vector data model
- Trajectories
- Spatial Tessellations
- Flows
- Practice: Python packages for geospatial analysis (Shapely, GeoPandas, folium, scikit-mobility)
Digital spatial and mobility data
- Mobile Phone Data
- GPS data
- Social media data
- Other data (POIs, Road Networks, etc.)
- Practice: reading and exploring spatial and mobility datasets in Python
Preprocessing mobility data
- filtering compression
- stop detection
- trajectory segmentation
- trajectory similarity and clustering
- Practice: data preprocessing with scikit-mobility

Module 2: Mobility Patterns and Laws

individual mobility laws/patterns
collective mobility laws/patterns
Practice: analyze mobility data with Python

Module 3: Predictive and Generative Models

Prediction
- Next-location prediction
- Crowd flow prediction
- Spatial interpolation
Generation
- Trajectory generation
- Flow generation
Practice: mobility prediction and generation in Python

Module 4: Applications

Epidemic spreading (COVID-19)
Urban segregation models
Routing and navigation apps
Traffic simulation with SUMO

Calendar

	Day	Topic	Slides/Code	Material	Teacher
1.	15.09 09:00-11:00	Introduction to the Course	[slides] Course modalities; [slides] Intro to Geospatial Analytics	[book chapter] Introduction to geographic information systems, Chapter 1; [paper] Human Mobility: Models and Applications, Section 1	Pappalardo, Nanni
2.	16.09 14:00-16:00	Fundamental Concepts - part 1	[slides] Fundamental concepts	[book chapter] Introduction to geographic information systems, Chapter 2 (Coordinate Systems); [paper] A survey of deep learning for human mobility, Section 2.1, Appendix A; Essentials of Geographic Information Systems,Chapter 4, Section 4.2 (Vector Data Models); [video] Intro to coordinate systems and UTM projection	Pappalardo
3.	22.09 11:00-13:00	Fundamental Concepts - part 2 (practice)	[code] Introduction to shapely, geopandas, folium, and scikit-mobility	[book chapter] Automating GIS-processes, Lesson 1 (Shapely and geometric objects); [article] Analyze Geospatial Data in Python: GeoPandas and Shapely; [paper] scikit-mobility: a Python library for the Analysis, Generation, and Risk Assessment of Mobility Data, Sections 1, 2;	Pappalardo
4.	23.09 09:00-11:00	Geographic and Mobility data - part 1	[slides] Geospatial and Mobility data	[paper] A survey of deep learning for human mobility , Appendix C.1, C.2, C.3; [paper] Evaluation of home detection algorithms on mobile phone data using individual-level ground truth , Sections “Introduction” and “Mobile phone datasets”; [paper] A survey of results on mobile phone datasets analysis , Sections “Introduction”, “Adding space - geographical networks”; [paper] Urban Human Mobility: Data-Driven Modeling and Prediction, Section 2.2;	Pappalardo
5.	29.09 11:00-13:00	Geographic and Mobility data - part 2 (practice)	[code] Geospatial and Mobility data in Python	[paper] scikit-mobility: a Python library for the Analysis, Generation, and Risk Assessment of Mobility Data, Section 4; [video] scikit-mobility data module; [tutorial] OSMnx: Python for Street Networks; [paper] OSMnx: New methods for acquiring, constructing, analyzing, and visualizing complex street networks; [book chapter] Intro to Python GIS, Retrieving OpenStreetMap data ;	Pappalardo
6.	06/10 11:00-13:00	Data preprocessing - part 1	[slides] Trajectory preprocessing	[paper] Review and classification of trajectory summarisation algorithms: From compression to segmentation; [paper] Algorithms for the reduction of the number of points required to represent a digitized line or its caricature (Douglas-Peucker); [paper] Approximating the Fréchet Distance for Realistic Curves in Near Linear Time (Driemel–HarPeled–Wenk); [paper] A Trajectory Segmentation Map-Matching Approach for Large-Scale, High-Resolution GPS Data; [paper] Hidden Markov Map Matching Through Noise and Sparseness	Nanni
7.	06/10 09:00-11:00	Data preprocessing - part 2	[slides] Semantic Enrichment	[paper] Analysis of human mobility patterns from GPS trajectories and contextual information; [paper] Using Mobile Positioning Data to Model Locations Meaningful to Users of Mobile Phones; [paper] Dynamic population mapping using mobile phone data; [paper] Inferring human activities from GPS tracks	Nanni
8.	13/10 11:00-13:00	Data preprocessing - part 3 (practice)	[code] Preprocessing Mobility data	[paper] scikit-mobility: a Python library for the Analysis, Generation, and Risk Assessment of Mobility Data, Section 3;	Cornacchia
9.	14/10 09:00-11:00	Individual Human Mobility Laws and Models - part 1	[slides] Individual mobility laws and models	[paper] The scaling laws of human travel; [paper] Understanding individual human mobility patterns; [paper] Human Mobility: Models and Applications, Sections 3.1 and 4; [paper] Returners and Explorers dichotomy in Human Mobility; [paper] Limits of predictability in human mobility; [paper] Modelling the scaling properties of human mobility;	Pappalardo
10.	20/10 11:00-13:00	Individual Human Mobility Laws and Models - part 2 (practice)	[code] Mobility laws and models	scikit-mobility documentation: measures, scikit-mobility documentation: models	Cornacchia
11.	21/10 09:00-11:00	Mobility Patterns	[slides] Mobility Patterns	[paper] A Survey on Trajectory Data Management, Analytics, and Learning, Section 3; [paper] Swarm: Mining Relaxed Temporal Moving Object Clusters; [paper] Computing longest duration flocks in trajectory data; [paper] Trajectory pattern mining; [paper] On Discovering Moving Clusters in Spatio-temporal Data	Nanni
12.	27/10 11:00-13:00	Collective Mobility Laws and Models - part 1	[slides] Collective mobility laws and models	[paper] Human Mobility: Models and Applications, Section 4.2; [paper] The P1 P2/D Hypothesis: On the Intercity Movement of Persons; [paper] Intervening Opportunities: A Theory Relating Mobility and Distance; [paper] A universal model for mobility and migration patterns; [paper] A tale of many cities: universal patterns in human urban mobility; [paper] Systematic comparison of trip distribution laws and models: [paper] A Deep Gravity model for mobility flows generation	Pappalardo
13.	28/10 09:00-11:00	Guest lecture	Joys and sorrows of mobile phone records		Leo Ferres
14.	03/11 11:00-13:00	Next-Location Prediction	[slides] Location prediction [code] Markov-chain-based location prediction	HMMlearn library; [paper] Mobility Prediction: A Survey on State-of-the-Art Schemes and Future Applications; [paper] A Survey on Trajectory-Prediction Methods for Autonomous Driving, Sections IV and V; [book chapter] Speech and Language Processing, Chapter A - Hidden Markov Models; [paper] Do Fielders Know Where to Go to Catch the Ball...?	Nanni
15.	04/11 09:00-11:00	Collective Mobility Laws and Models - part 2 (practice)	[code] Flow generation	scikit-mobility documentation: Gravity; scikit-mobility documentation: Radiation	Pappalardo
16.	10/11 11:00-13:00	Spatial segregation models - part 1	[slides] Segregation models	[paper] Dynamic models of segregation, Schelling; [paper] Segregation in networks; [paper] Mobility patterns are associated with experienced income segregation in large US cities;	Mauro
17.	11/11 09:00-11:00	Spatial segregation models - part 2 (practice)	[code] Implementing the Schelling model with MESA	[tutorial] Introduction to MESA	Mauro, Gambetta
18.	17/11 11:00-13:00	Traffic Simulation with SUMO	[slides] Traffic simulation with SUMO; [code] Traffic simulation with SUMO		Cornacchia
19.	18/11 09:00-11:00	Routing on road networks	[code] Routing on road networks		Cornacchia
20.	24/11 11:00-13:00	Presentation of projects/1	List of projects		Pappalardo, Nanni, Cornacchia, Mauro
21.	25/11 9:00-11:00	Presentation of projects/2	List of projects		Pappalardo, Nanni, Cornacchia, Mauro
22.	01/12 11:00-13:00	Guest Lecture			Fabrizio Martini from Electra Vehicles

Exam dates

The exam can be done on one of the following dates:

January 27th, 2023
February 24th, 2023

The exam will start at 9:30 am and will be in Aula Faedo (C-29) at ISTI-CNR, Pisa. Remember to bring an identity document (mandatory) and your “libretto” (if any).

Choose one of the two dates and remember that the project material must be submitted 10 days before the chosen date (i.e., January 17th and February 14th) through this google form.

The exam will consist of a discussion of the project and some questions about the course topics related to the project. The discussion of the project consists of a presentation by the student of the submitted notebook(s).