Pervasive data science
Individual course
Max amount of FITech students: 10
Persons without a valid study right at a Finnish university or university of applied sciences have preference to this course.
The course provides an advanced understanding of Pervasive Data Science (PDS) as an emerging research field resulting from the intersection of data science and pervasive computing. The course helps the students to learn the fundamentals, architecture and deployment of pervasive data science systems and deepen their knowledge about the applications and emerging challenges and opportunities of PDS.
The course provides a context for understanding how the knowledge obtained in diverse areas relates to pervasive data science and vice-versa. The development of further competencies depends on personal focus, as such:
- Sensing: data science, mobile sensing, machine learning, internet of things, big data management
- Pervasive computing: cloud and edge computing, distributed systems, advanced networking courses
Course contents
Topic 1: Introduction
- Pervasive data science: What is PDS?, applications and current direction
- The sensing pipeline
Topic 2: Data collection and feature engineering
- Ground truth
- Signal processing
- PDS measurements
Topic 3: Modelling and evaluation
- Data modelling: constraints for modelling, clustering
- Evaluation: model performance, model evaluation, robustness
- Deep learning and federated learning
Topic 4: PDS programming
- Programming considerations for PDS applications
- Considerations to implement deep learning and federated learning applications in Python
Topic 5: Other topics
- Privacy and security
Course material
The course does not follow any certain course book or set of papers, but each lecture is prepared individually. The teaching materials give references for further readings at the end of each lecture slide set.
The exercises are performed on Python and Jupyter notebooks.
Teaching schedule
Lectures on Tuesdays and Thursdays at 16-18.
Completion methods
The course doesn’t have an exam. The completion of the course is primarily based on a project work and the weekly tasks (around 50 %). There are no attendance requirements, but active participation during the sessions will be taken into account.
More information in the University of Helsinki study guide.
You can get a digital badge after completing this course.
IoT esineiden internet asioiden internet tekoäly koneoppiminen ML hajautettu oppiminen syväoppiminen tietojenkäsittelytiede
Responsible teacher
Further information about the course and studying
Contact person for applications
AI and machine learning,
Smart systems
Degree student