Postdoctoral Researcher · Cardiff University

John Stewart
Fabila-Carrasco

Graphs, signals, and machine learning.

I work on NLP and graph representation learning at Cardiff University. My research spans graph-based learning, scalable clustering and biomedical signals, with reproducible experiments in Python and PyTorch.

Research Publications CV (PDF)Contact

15Peer-reviewed papers

207Google Scholar citations

23Research talks

250+Teaching contact hours

Research interests

Research overview

NLP and graph representation learning

NLP methods that use graph representations and reasoning, with reproducible PyTorch experiments.

Spectral graph methods

Spectral tools for graph comparison, community detection and constrained clustering.

Graph signals and scientific data

Entropy and complexity measures for EEG, fMRI and other signals on networks.

Public-sector analytics

Statistical models, GIS maps and R/Shiny dashboards for national education data.

Research appointments

Sep 2025 - Apr 2027
Cardiff UniversityPostdoctoral Researcher in NLP and Graph Representation Learning
Feb 2024 - Apr 2025
University of Edinburgh, School of InformaticsPostdoctoral Researcher in Scalable Spectral Clustering
Nov 2020 - Jan 2024
University of Edinburgh, School of EngineeringPostdoctoral Researcher in Graph Signal Processing for Biomedical Data
Feb 2014 - Sep 2015
National Institute of Educational Evaluation (INEE)Industry and Government Data Science Researcher

Full research experience

Education

2020
PhD in Mathematical EngineeringUniversidad Carlos III de Madrid
2016
MSc in Mathematical EngineeringUniversidad Carlos III de Madrid
2012
MSc in Mathematical SciencesNational Autonomous University of Mexico
2010
BSc in MathematicsAutonomous Mexico State University

Education and distinctions

Selected publications

All publications

2025
Signed Laplacians for Constrained Graph Clustering
J. S. Fabila-Carrasco, H. Sun
International Conference on Machine Learning (ICML), 2025
Paper PDF
2024
Isospectral Graphs via Spectral Bracketing
J. S. Fabila-Carrasco, F. Lledo, O. Post
Linear Algebra and its Applications, 2024
DOI
2024
Graph-based Permutation Patterns for the Analysis of Task-related fMRI Signals on DTI Networks in Mild Cognitive Impairment
J. S. Fabila-Carrasco, A. Campbell-Cousins, M. A. Parra-Rodriguez, J. Escudero
IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
DOI
2022
Permutation Entropy for Graph Signals
J. S. Fabila-Carrasco, C. Tan, J. Escudero
IEEE Transactions on Signal and Information Processing over Networks, 2022
DOI Paper

Talks

23 research talks across nine countries, in graph theory, signal processing, and network science.

Talk record Seminar topics

Teaching & supervision

250+ contact hours, three MSc co-supervisions, and a Teaching Excellence Award.

Courses and supervision

Contact

For correspondence and other enquiries, please get in touch.

Fabila-CarrascoJ@cardiff.ac.uk

Get in touch LinkedIn Experience & CV