Recommended prior knowledge:
Programming experience with Python; Familiarity with machine learning and data minig; Familiarity with Deep Learning (have already taken the course or taking it parallel)
Urban computing, as an interdisciplinary field, is the science of using computing technology in solving urban challenges such as traffic, and pollution. This course provides a comprehensive introduction to this topic. Throughout the course, students will get familiar with data acquisitions, integration, and modeling skills required for urban computing research.
Spatio-temporal data is the data collected as a result of movement of mobile objects, such as cars and people, over space. Spatio-temporal data mining is the core to most urban computing research projects. Most previously designed machine learning algorithms do not consider the specific characteristics of spatio-temporal data. In this course, we will talk about how different classes of machine learning algorithms have been designed so far to perform on such data. We will also learn how additional data sources can be fused and integrated with spatio-temporal data. The concepts throughout the course are explained using examples extracted from the state of the art urban computing research literature. Students will put the knowledge of spatio-temporal data mining they acquire in this course into practice, using different assignments and a final project.
This course aims at providing knowledge on computational skills for urban computing research. Topics to be addressed are:
Acquiring data for urban computing research
Processing time-series data
Processing spatial data
Processing spatio-temporal data
Visualization techniques for spatio-temporal data
Machine learning algorithms for urban computing research
Deep learning for urban computing research
The most recent timetable can be found at the Computer Science (MSc) student website.
More detailed information can be found at Brightspace.
Mode of instruction
Total hours of study: 168 hrs. (= 6 EC)
Lecture hours (lectures + presentations): 24 hrs.
Practical assignments: 40 hrs.
Final project: 104 hrs.
The final score of the course is composed of:
Presentation (15 %)
Project peer review (10%)
It is possible to resit assignments and the project. However, the resit of assignments gets a maximum grade of 7. To pass the course the course combined grade, project grade and each assignments grade should be over 5.5. No resit option is available for presentations. Submitting a project proposal is compulsory (no submission implies deduction of 20% of the project grade).
- Extra reading material composed of recent papers will become available through Brightspace.
- You have to sign up for courses and exams (including retakes) in uSis. Check this link for information about how to register for courses.
Lecturer: dr. M. Baratchi