Graham McDonald

Lecturer in Information Retrieval at the University of Glasgow

About Me

View Google Scholar profile here
ORCID 0000-0002-1266-5996

I am a Senior Lecturer in Information Retrieval (IR) in the Information, Data and Analysis (IDA) Section of the School of Computing Science, University of Glasgow. Working within the Terrier Team, I am PI on the EPSRC funded project Safe Information Extraction from Patient Histories (SIEPH) working with NHS SafeHavens, I was General co-Chair of the 46th European Conference on Information Retrieval (ECIR 2024) and I am the Vice Chair of the BCS Information Retrieval specialist group. I was awarded my PhD in information retrieval from the University of Glasgow in 2019 for my thesis entitled A Framework for Technology-Assisted Sensitivity Review: Using Sensitivity Classification to Prioritise Documents for Review, under the supervision of Prof. Iadh Ounis and Dr. Craig Macdonald. I previously attained a MSci in Computing Science from the University of Glasgow in 2013 and, subsequently, have held three full-time Research Assistant positions within the school.

Research Interests

My main research interests are in responsible information retrieval and text classification. In particular, my research focuses on automatically classifying sensitive information that is exempt from public release through freedom of information laws, assisting and learning from human reviewers in active learning scenarios for technology-assisted reviews, and generating fair ranking strategies that provide an appropriate level of exposure to relevant information producers in search results over a period of time.

I am part of a consortium lead by SVGC that is tasked with delivering the first large-scale technology-assisted digital sensitivity review platform for a major central UK government department. This system is the first of its kind within the UK. My role within the consortium is to research and develop machine learning-based search and classification technologies to assist sensitivity review by automatically identifying information that is likely to be exempt from public release under the Freedom of Information Act (2000).

I am an organiser of the Text REtrieval Conference (TREC) Fair Ranking Track. TREC is an international evaluation forum that facilitates research groups to develop novel solutions to cutting edge problems in information retrieval, and is responsible for developing many of the core scientific advances that drive today’s information retrieval landscape. The Fair Ranking Track 2021 code and resources are available here.

Tutorial on Search Among Sensitive Content

European Conference on Inforamtion Retrieval (ECIR 2021)

Resources and full video recording of the tutorial are avaiable here.

Fairness and Bias in Information Retrieval Workshop

University of Glasgow (Online)

I recently organised a full-day online workshop on Fairness and Bias in Information Retrieval (#FAIRER2021). The workshop brought together practitioners from academia and industry to discuss the challenges relating to fairness in IR that are faced by industry, recent advances in fair IR research, and new research directions for creating fair IR systems. Video presentations from the workshop are available here and a review of the workshop can be found here.

Publications

PhD Thesis

• A Framework for Technology-Assisted Sensitivity Review: Using Sensitivity Classification to Prioritise Documents for Review
  Graham McDonald, University of Glasgow, 2019 (PDF)

Publications

• Search Results Diversification for Effective Fair Ranking in Academic Search
  Graham McDonald, Craig Macdonald and Iadh Ounis. Information Retrieval Journal (IRJ) Dec 2021, pp. 1-26 (PDF)
• RelDiff: Enriching Knowledge Graph Relation Representations for Sensitivity Classification
  Hitarth Narvala, Graham McDonald and Iadh Ounis. Findings of the Association for Computational Linguistics (EMNLP 2021) (PDF)
• How the Accuracy and Confidence of Sensitivity Classification Affects Digital Sensitivity Review
  Graham McDonald, Craig Macdonald and Iadh Ounis. ACM Transactions on Information Systems (TOIS) 39 (1), 1-34 (PDF)
• Active Learning Stopping Strategies for Technology-Assisted Sensitivity Review
  Graham McDonald, Craig Macdonald and Iadh Ounis. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval 2020, (SIGIR 2020) (PDF)
• Receptor: A Platform for Exploring Latent Relations in Sensitive Documents
  Hitarth Narvala, Graham McDonald and Iadh Ounis. In Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval 2020, (SIGIR 2020) (PDF)
• FACTS-IR: Fairness, Accountability, Confidentiality, Transparency, and Safety in Information Retrieval
  Roegiest et al. In ACM SIGIR Forum, Vol. 53, pp 20-43, (FACTS-IR @ SIGIR) (PDF)
• The FACTS of Technology-Assisted Sensitivity Review
  Graham McDonald, Craig Macdonald and Iadh Ounis. In the 1st FACTS-IR workshop @ SIGIR 2019, (FACTS-IR 2019)(PDF)
• How Sensitivity Classification Effectiveness Impacts Reviewers in Technology-Assisted Sensitivity Review
  Graham McDonald, Craig Macdonald and Iadh Ounis. In Proceedings of the ACM SIGIR Conference on Human Information Interaction and Retrieval 2019, (CHIIR 2019) (PDF)
• Active Learning Strategies for Technology Assisted Sensitivity Review
  Graham McDonald, Craig Macdonald and Iadh Ounis. In Proceedings of the 40th European Conference on Information Retrieval 2018, (ECIR 2018) (PDF)
• Towards Maximising Openness in Digital Sensitivity Review using Reviewing Time Predictions
  Graham McDonald, Craig Macdonald and Iadh Ounis. In Proceedings of the 40th European Conference on Information Retrieval 2018, (ECIR 2018) (PDF)
• A Study of SVM Kernel Functions for Sensitivity Classification Ensembles with POS Sequences
  Graham McDonald, Nicolás García-Pedrajas, Craig Macdonald and Iadh Ounis. In Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval 2017, (SIGIR 2017) (PDF)
• Enhancing Sensitivity Classification with Semantic Features using Word Embeddings
  Graham McDonald, Craig Macdonald and Iadh Ounis. In Proceedings of the 39th European Conference on Information Retrieval 2017, (ECIR 2017) (PDF)
• Using Part-of-Speech N-grams for Sensitive-Text Classification
  Graham McDonald, Craig Macdonald and Iadh Ounis. In Proceedings of the ACM SIGIR International Conference on The Theory of Information Retrieval 2015, (ICTIR 2015) (PDF)
• A Framework for Enhanced Text Classification in Sensitivity and Reputation Management
  Graham McDonald. The 6th (BCS-IRSG) Symposium on Future Directions in Information Access 2015, (FDIA 2015) (PDF)
• Tweet Enrichment for Effective Dimensions Classification in Online Reputation Management
  Graham McDonald, Romain Deveaud, Richard McCreadie, Craig Macdonald and Iadh Ounis. In Proceedings of the 9th International Conference on Web and Social Media 2015, (ICWSM 2015) (PDF)
• On Using Information Retrieval for the Selection and Sensitivity Review of Digital Public Records
  Timothy Gollins, Graham McDonald, Craig Macdonald, and Iadh Ounis. In Proceedings of Privacy-Preserving IR: When Information Retrieval Meets Privacy and Security, workshop @ SIGIR 2014, (PIR 2014) (PDF)
• Towards a Classifier for Digital Sensitivity Review
  Graham McDonald, Craig Macdonald, Iadh Ounis, and Timothy Gollins. In Proceedings of the 36th European Conference on Information Retrieval 2014, (ECIR 2014) (PDF)

International Evaluation Forums

• University of Glasgow Terrier Team and Naver Labs Europe at TREC 2019 Fair Ranking Track
  Graham McDonald, Thibaut Thonet, Iadh Ounis, Jean-Michel Renders and Craig Macdonald. In Proceedings of the Text REtrieval Conference 2019, (TREC 2019). (PDF)
• University of Glasgow at TREC 2015: Experiments with Terrier in Contextual Suggestion, Temporal Summarisation and Dynamic Domain Tracks
  Richard McCreadie, Stuart Mackie, Jarana Manotumruksa, Graham McDonald, Saul Vargas, M-Dyaa Albakour, Craig Macdonald and Iadh Ounis. In Proceedings of the Text REtrieval Conference 2015, (TREC 2015). (PDF)
• University of Glasgow Terrier Team / Project Abaca at RepLab 2014: Reputation Dimensions Task
  Graham McDonald, Romain Deveaud, Craig Macdonald, Richard McCreadie, Timothy Gollins and Iadh Ounis. In Proceedings of CLEF 2014: Conference and Labs of the Evaluation Forum 2014, (CLEF 2014) (PDF)
• University of Glasgow at Medical Records Track 2011: Experiments with Terrier
  Nut Limsopatham, Craig Macdonald, Iadh Ounis, Graham McDonald, and Matt-Mouley Bouamrane. In Proceedings of the Text REtrieval Conference 2011, (TREC 2011). (PDF)

Activities

Invited Talks

• For Your Eyes Only: Scalable Approaches to Identifying, Managing, and Securing Sensitive or Restricted Informarion, Digital Preservation Coalition, "A Technical Framework for Automating the Detection of Sensitive Data", 2020
• Subject Matter Experts Meeting, NATO - North Atlantic Treaty Organization, "Technology-Assisted Sensitivity Review", 2018
• Machine Learning for Information Management, Université Libre de Bruxelles, "Using ML to assist the discovery of sensitive information in documents", 2018
• Information Management Liaison Group, The National Archives, "Supervised Classification and Presentation Strategies for Assisted Sensitivity Review", 2018
• Xerox Research Centre Europe, "Sensitivity Classification for Assisting Sensitivity Review of Government Documents", 2017
• Archives Without Borders Workshop, The National Archives, "An Ensemble Classification Approach Towards Identifying Sensitive Information in Documents", 2017

Committees

• ECIR, Short Paper Track, 2020
• ACM SIGIR, Demo Track, 2018
• WWW - The Web Conference, Poster Track, 2018
• University of Glasgow Research Postgraduate Event: The Future of Computational Science, 2017
• The Scottish Informatics and Computer Science Alliance (SICSA), PhD Conference, 2015, 2016 and 2017

Contact

Graham McDonald
406 Sir Alwyn Williams Building
School of Computing Science
University of Glasgow
Glasgow
G12 8QQ

Email: Graham.McDonald@Glasgow.ac.uk
Twitter: @graham_mcdonald
Telephone: +44(0) 141 330 5458

Copyright © Graham McDonald 2016. Adapted from a Start Bootstrap theme.