Introduction to network analytics with Python (01/10/2024)
Network analytics leverages the relations among entities, next to their attributes, to come to a more exhaustive analysis. It has found numerous applications, as networks form a natural framework for many problems. Also in actuarial science, be it life, non-life or health, we can extend our current notion of the work using networks.
During this presentation, we are quickly going to go over some basic notions about networks and network representations. Then, we are going to take a deep dive into different methods used to extract network information to apply it for machine learning problems. This will range from manual feature engineering to capture local features, to state-of-the-art graph neural networks, based on deep learning.
To have a good balance between theory and practice, material will be provided on the implementation of these methods in the Python programming language. We are going to discuss an example based on an open-source data set during the training.
After this training, the participant should have an understanding of different network analytics methods and be able to train and run these methods on their own data sets.
Although the topic of this training is quite technical, no previous knowledge is required. A basic working knowledge of Python is preferred to be able to easily follow along during the coding examples.
Lecturer: Bruno Deprez, PhD Researcher - KU Leuven LIRIS - UAntwerp Department of Mathematics, Winner Second IA|BE Prize 2023
Some practical information:
The training will take place using TEAMS. The link will be provided to the participants the day before the training will take place.
In case of any question, please contact us.