Multi-relational Graph Modeling and Learning for Large-Scale Social Network Analysis

Offer Description

The project proposes a theoretical and
methodological study on multi-relational modeling of large-scale social networks using graph representation
learning techniques. The objective is to investigate how different interaction types (e.g., user relations, content
usage, temporal dynamics) can be modeled as distinct relational views of the same network and integrated into a
unified framework. The research relies on a real-world dataset covering the entire Italian Twitter sphere in 2022
and aims to evaluate modular Graph Neural Network architectures for scalable and interpretable node
representation learning. Applications to link prediction, community detection, and emerging social dynamics
analysis will be explored

• Italy

Eligibility of fellows: country/ies of residence:

• OTHER

Eligibility of fellows: nationality/ies:

#J-18808-Ljbffr