Computer Science - Benchmarking Temporal k-Core Algorithms

Supervisor: Dr Lutz Oettershagen

Supervisor bio: 

My primary research areas are algorithmic data analysis, graph data mining, and machine learning for graphs. I put a strong focus on mathematical and computational foundations as well as the engineering and application of efficient algorithmic data analysis on (dynamic) graphs for solving real-world problems. My work focuses on the computational analysis of static and temporal networks. In my research, I design and analyze methods for obtaining new knowledge from complex networks.

Until the end of June 2024, I was a postdoctoral researcher at the KTH Royal Institute of Technology, Stockholm, Sweden, working with Prof. Dr. Aristides Gionis. Before that, I was a postdoctoral researcher at the Lamarr Institut and the University of Bonn in Bonn, Germany, working with Prof. Dr. Petra Mutzel.

Email: lutz.oettershagen@liverpool.ac.uk

School: School of Computer Science and Informatics 

Department: Computer Science

Module code: COMP298

Suitable for students of Computer Science

Desirable experience/requirements: 

Basic programming experience (Python or  C++). Interest in networks, graph algorithms, or data analysis.

Places available: 2

Start date: Session 1 (15th June 2026),Session 2 (6th July 2026)

Project length: 4 or 8 weeks 

Virtual option: Yes

Hybrid option: Yes

Project description: 

Temporal networks capture how interactions between entities evolve over time, and temporal k-core decomposition provides a powerful tool for identifying dense, persistent substructures within these dynamic graphs. Multiple definitions and algorithms exist for computing temporal k-cores, but their performance on large and real-world datasets is not well understood.

This project focuses on benchmarking several temporal k-core algorithms—both classical and recent approaches—on datasets such as mobility traces, communication networks, social interactions, or synthetic dynamic graphs. Students will:

- Implement or adapt existing temporal k-core algorithms.
- Evaluate their performance (runtime, memory, scalability) across varying temporal resolutions.
- Analyse how algorithmic choices affect the structural insights obtained from the data.
- Produce visualisations and a research-style summary/poster of findings.

The project offers the opportunity to explore real-world data, algorithm engineering, and temporal graph analysis, with flexibility to emphasise coding, data science, or interpretation depending on the student’s background and preferences.

Additional requirements: N/A