Unsupervised Learning on Social Data

MCML Authors

Felix Borutta

Dr.

* Former Member

→ Group Thomas Seidl
 Database Systems, Data Mining and AI

Abstract

This thesis addresses several challenges in social data analytics, focusing on methods for clustering, learning from network data, and analyzing dynamic social data. It introduces novel algorithms for correlation clustering on streaming data, hierarchical clustering for social maps, and user identification based on spatio-temporal mobility patterns. Additionally, the thesis presents various node embedding techniques for learning representations from network topology and proposes a graph neural network model for matching nodes across overlapping graphs. (Shortened.)

phdthesis

---

Dissertation

LMU München. Mar. 2020

---

Authors

F. Borutta

---

Links

DOI

---

Research Area

 A3 | Computational Models