30422 - TECHNOLOGICAL INNOVATION SEMINARS II
Department of Decision Sciences
OMIROS PAPASPILIOPOULOS
Mission & Content Summary
MISSION
CONTENT SUMMARY
- Introduction to networks
- Network statistics and connectivity properties
- Visualization and community detection (force-directed placement algorithms, community detection algorithms)
- Basic models for networks (stochastic block models, latent space models)
Intended Learning Outcomes (ILO)
KNOWLEDGE AND UNDERSTANDING
- Distinguish between model-based and algorithmic-based techniques for the analysis of network data
- Describe the main properties and implementation details of model-based and algorithmic-based techniques for the analysis of network data
- Identify the most suitable technique to analyse a given network according to a specific research objective
APPLYING KNOWLEDGE AND UNDERSTANDING
- Apply standard statistical softwares for the analysis of network data
- Develop codes to implement specific model-based and algorithmic-based analyses of networks
- Discuss the ouput of relevant statistical models for network data
Teaching methods
- Lectures
DETAILS
Each method presented in the face-to-face lectures is directly motivated and illustrated on a number of relevant case studies from political sciences, neurosciences and criminology. These case studies showcase the potentials of Data Science, Machine Learning and AI in the specific field of Network Science.
Assessment methods
Continuous assessment | Partial exams | General exam | |
---|---|---|---|
|
x | ||
|
x |
ATTENDING STUDENTS
Full credit is assigned automatically to all students who meet the active class participation condition (3 or 4 classes with registered attandance out the 4 total classes). These students do not need to take the oral exam.
NOT ATTENDING STUDENTS
For students who do not meet the active class participation condition, the full credit is assigned after a successful oral individual exam on the topics presented during the course.
In particular, the oral individual exam will be based on questions assessing the ability of students to distinguish between model-based and algorithmic-based techniques for the analysis of network data, while describing the main properties, implementation details and outputs of these techniques.
Teaching materials
ATTENDING AND NOT ATTENDING STUDENTS
The course is entirely based on slides and research articles. All slides and research articles are made available to both attending and non-attending students.