Dr Evdoxia Taka

Research Associate

University of Glasgow UK

I am a Postdoctoral Researcher in the School of Computing Science at the University of Glasgow. My research currently focuses on responsible (probabilistic) AI in Criminal Justice.

My degrees include:

  • Diplom (MSc. equiv.) in Electrical & Computer Engineering, Aristotle University of Thessaloniki, Greece
  • MSc. in Digital Media & Computational Intelligence, Aristotle University of Thessaloniki, Greece
  • Ph.D. in Computer Science, University of Glasgow, UK

Postdoctoral Experience (School of Computing Science, University of Glasgow):

  • May 2023 - Jan 2023: EFFI: End-users Fixing Fairness Issues. Funded by Fujitsu Ltd
  • Jan 2023 - Jan 2024: Women, Ageing, and Machine Learning on Screen. Funded by the Leverhulme Institute
  • Jan 2024 - Present: PROBabLE Futures. Funded by RAI UK

My research interests include:

  • Uncertainty Visualization
  • Bayesian Probabilistic Modelling
  • Explainable AI
  • Fairness in AI
  • Human-in-the-loop approaches
  • Machine Learning
  • Causality

Research

Postdoctoral Researcher

School of Computing Science, University of Glasgow, UK
Jan. 2025 - Present
PROBabLE Futures. Funded by RAI UK.
  • E. Taka, T. Lawal, K. Jones, M. Sevegnani, M. Oswald, M. Calder, Panel Discussion: "Probabilistic AI in Criminal Justice: Mapping the Ecosystem and Scoping for the Future", UK AI Research Symposium 2025, Oct 2025, proposal
  • E. Taka, T. Lawal, K. Jones, "Mapping the AI Ecosystem in UK Criminal Justice System and Participatory Activities", Scottish International Policing Conference (SIPC) 2025, Organizer: Scottish Institute for Policing Research (SIPR), May 2025
  • E. Taka, T. Lawal, M. Calder, M. Sevegnani, K. Kotsoglou, E. McClory-Tiarks, M. Oswald, "Mapping the Probabilistic AI Ecosystem in Criminal Justice in England and Wales", paper under preparation
  • M. Calder, M. Oswald, E. McClory-Tiarks, M. Sevegnani, E. Taka, "Responsible AI in Criminal Justice: potential use of LLMs in policing and risks to case progression", paper under preparation
Jan. 2024 - Jan. 2025
Women, Ageing, and Machine Learning on Screen. Funded by the Leverhulme Trust.
  • E. Taka, D. Bhattacharya, J. Garde-Hansen, S. Sharma, T. Guha, "Analyzing Character Representation in Media Content using Multimodal Foundation Model: Effectiveness and Trust", ICMI '25: Proceedings of the 27th International Conference on Multimodal Interaction, 2025, doi: 10.1145/3716553.3750785
May 2023 - Jan. 2024
EFFI: End-users Fixing Fairness Issues. Funded by Fujitsu Ltd.
  • E. Taka, Y. Nakao, R. Sonoda, T. Yokota, L. Luo, S. Stumpf, "Human-in-the-loop Fairness: Integrating Stakeholder Feedback to Incorporate Fairness Perspectives in Responsible AI", 2024, url: arXiv.2312.08064 [cs.AI].
  • S. Stumpf, E. Taka, Y. Nakao, L. Luo, R. Sonoda, T. Yokota, "The Need for User-centred Assessment of AI Fairness and Correctness", UMAP Adjunct '24: Adjunct Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization, 2024, doi: 10.1145/3631700.3664912
  • S.Stumpf, E. Taka, Talk: "End-user Fixing Fairness Issues (EFFI) - The Impact of Lay User Feedback for Improving AI Fairness", GIST Seminars, Oct 2023
Apr 2019 - Apr 2023

PhD

School of Computing Science, University of Glasgow, UK
Thesis: Interactive Animated Visualization of Probabilistic Models, 2023, doi: 10.5525/gla.thesis.83903

Scholarship by the EPSRC Closed-Loop Data Science for Complex, Computationally- and Data-Intensive Analytics project

  • E. Taka, S. Stein, J. H. Williamson, "Increasing Interpretability of Bayesian Probabilistic Programming Models Through Interactive Representations", Frontiers in Computer Science Journal, 2020, doi: 10.3389/fcomp.2020.567344
  • E. Taka, S. Stein, J. H. Williamson, "Does Interactive Conditioning Help Users Better Understand the Structure of Probabilistic Models?", IEEE Transactions on Visualization and Computer Graphics, 2023, doi: 10.1109/TVCG.2022.3231967
  • E. Taka, S. Stein, J. H. Williamson, Dataset: "Does Interacting Help Users Better Understand the Structure of Probabilistic Models?", University of Glasgow, 2022, doi: 10.5525/gla.researchdata.1248
  • E. Taka, S. Stein, J. H. Williamson, Poster: "Closed-Loop Interactions With Probabilistic Models", SICSA PhD Conference 2019
  • E. Taka, Presentation: "Automatic Transformation of Bayesian Probabilistic Models Into Interactive Visualizations", PYMCON 2020
  • Visualization Tools:
    • IPME (Interactive Probabilistic Models Explorer) (github)
    • IPP (Interactive Pair Plot) (github)
    • vicausi (Visualizer of Causal Assumptions and Uncertainty-Aware Simulations of Interventions) (github)
  • Probabilistic Simulator of Interventions (github)
Jan 2018 - Apr 2019

Research Database Engineer

Centre for Virus Research (CVR), University of Glasgow, UK
HCV Research UK Project
  • Creation of a web application for integrating, accessing, and managing the clinical & research data from the HCV research UK project
Jun 2014 - Aug 2017

Research Assistant

Information Technologies Institute (ITI), Greece
IPv_Park: Intelligent system for PhotoVoltaic PARK management, National Funding
  • Gait Analysis
P-REACT: Petty cRiminality diminution through sEarch and Analysis in multi-source video Capturing and archiving plaTform, FP7-SEC-2013
  • N. Dimitriou, G. Kioumourtzis, A. Sideris, G. Stavropoulos, E. Taka, N. Zotos, G. Leventakis, D. Tzovaras, "An Integrated Framework for the Timely Detection of Petty Crimes", IEEE EISIC Conference 2017, doi: 10.1109/EISIC.2017.13
Scan4Reco: Multimodal Scanning of Cultural Heritage Assets for their multilayered digitization and preventive conservation via spatiotemporal 4D Reconstruction and 3D Printing, Horizon-2020
  • E. Taka, K. Papachristou, A. Drosou, N. Dimitriou, D. Tzovaras, "Physical Forces aware of Aging Simulation on Cultural Heritage Artifacts", IEEE 3DTV Conference 2017, doi: 10.1109/3DTV.2017.8280394
  • E. Taka, K. Papachristou, N. Dimitriou, A. Drosou, D. Tzovaras, "On the potential of Simulation enhanced Conservation of CH Artifacts", IEEE EEEIC Conference 2017, doi: 10.1109/EEEIC.2017.7977688

Teaching

AY 21/22

Graduate Teaching Assistant

School of Computing Science, Glasgow University, UK
  • Data Fundamentals (H) - COMPSCI4073
  • Machine Learning and Artificial Intelligence for Data Scientists (M) - COMPSCI5100
  • Human - Computer Interaction (H) - COMPSCI4023
AY 20/21

Graduate Teaching Assistant

School of Computing Science, Glasgow University, UK
  • Data Fundamentals (H) - COMPSCI4073
  • Systems Programming (H) - COMPSCI4081
  • Introduction to Data Science and Systems (M) - COMPSCI5089
  • Deep Learning (M) - COMPSCI5085
  • Object-Oriented Software Engineering 2 - COMPSCI2008
AY 19/20

Graduate Teaching Assistant

School of Computing Science, Glasgow University, UK
  • Information Retrieval (H/M) - COMPSCI5011 (only marking support)

Supervision

Master Projects
  • Debadyuti Bhattacharya, Visualizing Diversity: An Automatic Deep-Learning Pipeline for Demographic Analysis in Media, August 2024. A dissertation presented in part fulfilment of the requirements of the Degree of Master of Science at The University of Glasgow. (led to publication)

Reviewer

Peer-reviewed Journals
  • Cruz D and Batista J (2024) Causality and tractable probabilistic models. Front. Comput. Sci. 5:1263386. doi: 10.3389/fcomp.2023.1263386
  • Yang M, El-Attar AA and Chaspari T (2024) Deconstructing demographic bias in speech-based machine learning models for digital health. Front. Digit. Health 6: 1351637. doi: 10.3389/fdgth.2024.1351637
Peer-reviewed Conferences
  • 27th ACM International Conference on Multimodal Interaction (ICMI) 2025
  • British Computer Society’s Human Computer Interaction Conference (BCS HCI) 2025

Research Outputs

Peer-reviewed Journals
  • E. Taka, S. Stein, J. H. Williamson, "Does Interactive Conditioning Help Users Better Understand the Structure of Probabilistic Models?", IEEE Transactions on Visualization and Computer Graphics, 2024, doi: 10.1109/TVCG.2022.3231967
  • E. Taka, S. Stein, J. H. Williamson, "Increasing Interpretability of Bayesian Probabilistic Programming Models Through Interactive Representations", Frontiers in Computer Science Journal, 2020, doi: 10.3389/fcomp.2020.567344
Peer-reviewed Conferences
  • E. Taka, D. Bhattacharya, J. Garde-Hansen, S. Sharma, T. Guha, "Analyzing Character Representation in Media Content using Multimodal Foundation Model: Effectiveness and Trust”, ICMI '25: Proceedings of the 27th ACM International Conference on Multimodal Interaction", 2025, doi: 10.1145/3716553.3750785
  • S. Stumpf, E. Taka, Y. Nakao, L. Luo, R. Sonoda, T. Yokota, "The Need for User-centred Assessment of AI Fairness and Correctness", UMAP Adjunct '24: Adjunct Proceedings of the 32nd ACM Conference on User Modeling, Adaptation and Personalization, 2024, doi: 10.1145/3631700.3664912
  • N. Dimitriou, G. Kioumourtzis, A. Sideris, G. Stavropoulos, E. Taka, N. Zotos, G. Leventakis, D. Tzovaras, "An Integrated Framework for the Timely Detection of Petty Crimes", IEEE EISIC Conference 2017, doi: 10.1109/EISIC.2017.13
  • E. Taka, K. Papachristou, A. Drosou, N. Dimitriou, D. Tzovaras, "Physical Forces aware of Aging Simulation on Cultural Heritage Artifacts", IEEE 3DTV Conference 2017, doi: 10.1109/3DTV.2017.8280394
  • E. Taka, K. Papachristou, N. Dimitriou, A. Drosou, D. Tzovaras, "On the potential of Simulation enhanced Conservation of CH Artifacts", IEEE EEEIC Conference 2017, doi: 10.1109/EEEIC.2017.7977688
Under Preparation/Review
  • E. Taka, S. Stein, J. H. Williamson, "Visualizing Simulated Interventions to Support Causal Reasoning"
  • E. Taka, T. Lawal, M. Calder, M. Sevegnani, K. Kotsoglou, E. McClory-Tiarks, M. Oswald, "Mapping the Probabilistic AI Ecosystem in Criminal Justice in England and Wales"
  • M. Calder, M. Oswald, E. McClory-Tiarks, M. Sevegnani, E. Taka, "Responsible AI in Criminal Justice: potential use of LLMs in policing and risks to case progression"
  • E. Taka, Y. Nakao, R. Sonoda, T. Yokota, L. Luo, S. Stumpf, "Human-in-the-loop Fairness: Integrating Stakeholder Feedback to Incorporate Fairness Perspectives in Responsible AI", url: arXiv.2312.08064 [cs.AI].
Datasets
  • E. Taka, D. Bhattacharya, J. Garde-Hansen, S. Sharma, T. Guha, "Analyzing Character Representation in Media Content using Multimodal Foundation Model: Effectiveness and Trust", 2025, github
  • E. Taka, Y. Nakao, R. Sonoda, T. Yokota, L. Luo, S. Stumpf: "Human participants' feedback on the fairness of AI loan application decisions in an Interactive Machine Learning setting with the system providing feedback", 2023, github
  • E. Taka, Y. Nakao, R. Sonoda, T. Yokota, L. Luo, S. Stumpf: "Human participants' feedback on the fairness of AI loan application decisions with the system not providing feedback", University of Glasgow, 2023, github
  • E. Taka, S. Stein, J. H. Williamson: "Evaluation of Decisions on Interventions using Interactive and Animated Uncertainty-Aware Visualizations of Simulated Data", 2023, github
  • E. Taka, S. Stein, J. H. Williamson, Dataset: "Does Interacting Help Users Better Understand the Structure of Probabilistic Models?", University of Glasgow, 2022, doi: 10.5525/gla.researchdata.1248
Theses
  • E. Taka, PhD Thesis: "Interactive Animated Visualization of Probabilistic Models", 2023, doi: 10.5525/gla.thesis.83903
  • E. Taka, Master's Thesis: "Dynamic Analysis Techniques for Recognizing Magneto-encephalographic (MEG) Connectivity Patterns During Spontaneous Blinking", 2016, doi: 10.26262/heal.auth.ir.285210
  • E. Taka, D. Antonaras, Master's Thesis: "Video Magnification for Revealing Subtle Changes in the Brain Blood Flow", 2014
Talks/Presentations
  • E. Taka, T. Lawal, K. Jones, M. Sevegnani, M. Oswald, M. Calder, Panel Discussion: "Probabilistic AI in Criminal Justice: Mapping the Ecosystem and Scoping for the Future", UK AI Research Symposium 2025, Oct 2025, proposal
  • E. Taka, T. Lawal, K. Jones, "Mapping the AI Ecosystem in UK Criminal Justice System and Participatory Activities", Scottish International Policing Conference (SIPC) 2025, Organizer: Scottish Institute for Policing Research (SIPR), May 2025
  • S.Stumpf, E. Taka, "End-user Fixing Fairness Issues (EFFI) - The Impact of Lay User Feedback for Improving AI Fairness", Seminar series organized by the GIST section of School of Computing Science, University of Glasgow , Oct 2023, video
  • E. Taka, "Automatic Transformation of Bayesian Probabilistic Models Into Interactive Visualizations", PYMCON 2020, video
  • E. Taka, S. Stein, J. H. Williamson, "Closed-Loop Interactions With Probabilistic Models", SICSA PhD Conference 2019, poster
Software/Tools
  • IPME (Interactive Probabilistic Models Explorer) (github)
  • IPP (Interactive Pair Plot) (github)
  • vicausi (Visualizer of Causal Assumptions and Uncertainty-Aware Simulations of Interventions) (github)
  • Probabilistic Simulator of Interventions (github)
  • A tool for the analysis and visualization of character representation in media by using foundation models (github)