Areas of Expertise

  • Machine Learning
  • Data Science
  • Wearables (EEG)
  • Applied Deep Learning
  • Signal Processing

Publications

  1. Roach, M., D'Cruze, T., Smith, G., Brookes, P., Laramee, R., Roberts, R., Roach, M. A Tale of Two Visions - Exploring the Dichotomy of Interest between Academia and Industry in Visualisation Proceedings of the 13th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications 3 IVAPP 319 326
  2. M.J, R., M, P., Roach, M. 'Image Characterisation' Patent number WO0239386 BT Technology Journal 20 4
  3. Pierre, M., Matthew, R., John, M., Roach, M. Visual Form 2001 Lecture notes in computer science
  4. M., R., J.S., M., M., P., Roach, M. Motion-based classification of cartoons Proceedings of 2001 International Symposium on Intelligent Multimedia, Video and Speech Processing. ISIMP 2001 (IEEE Cat. No.01EX489)
  5. M.J., R., J.D., B., J.S.D., M., Roach, M. Acoustic and facial features for speaker recognition Proceedings 15th International Conference on Pattern Recognition. ICPR-2000

See more...

Teaching

  • CSC005 Computational Probability

    This module gives students an understanding of Monte Carlo methods, which use random simulation to computationally solve physical problems which may be difficult if not impossible to solve analytically. The delivery style of the module is very much tutorial in nature: only basic programming and probability theory will be taught and required, with the module concentrating on practical weekly lab-based problem-solving sessions.

  • CSCM10 Computer Science Project Research Methods

    This module will introduce students to some fundamental research methodologies and good practice in research. They will undertake background research including a literature review and specify the aims of their MSc project.

  • CSCM21 Abuses, Biases and Blessings of Data

    This module will take a look at how Machine Learning and Artificial Intelligence (black box) and data driven systems are applied in various real-world contexts. It will cover a variety of successful and unsuccessful case studies from academia and industry. Lectures will provide theoretical framework through which students will learn to construct their analysis and evaluate various forms of ML / AI algorithms for their appropriateness to be deployed in society as decision support systems. Students will be expected to read published literature and contribute to seminars by preparing presentations on new material and take part in reasoned scientific debate on contemporary ethical issues in the design and deployment of data and Intelligent systems.

  • CSDM001 Theory of Digital Information and Society

    This module gives an overview on theoretical and methodological debates contemporary Digital Economy and Society research with specific focus on Human Computer Interaction. This module helps students understand how human experience can influence the design and adoption of digital into services and the lives of individuals and communities.Students explore the advanced literature and research results underpinning the field. Understand, through a series of Classic papers, the practical application of qualitative and quantitative techniques for the study of soci-technical assemblages in a digital by default world, as well as recent work from the leading figures. Students achieve a clear view of the 'cutting edge' and issues in the field and where things are happening. The module is very interactive, and students will be expected to give presentations.

  • CSDM03 Computational Thinking skills for Digital Social Scientists.

    This module will discuss some of the most widely used and artificial intelligence and machine learning, regression & clustering techniques and their applications to big data social science questions. The students will gain and understanding of both strengths and weaknesses of learning and practical know-how in applying those theories to real world problems. Topics include big social data concepts, data mining, learning theories, supervised and unsupervised learning, and reinforcement learning.

Supervision

  • Predicting Outcomes within a Rugby match using machine learning approaches (current)

    Student name:
    MSc
    Other supervisor: Prof Xianghua Xie
  • "Following the SWP Uniform": A Play With 'Bleeding Humans' (current)

    Student name:
    PhD
    Other supervisor: Dr Sergei Shubin
  • Rethinking sustained healthy lifestyle change through design and evaluation of socio-technical wearable systems (current)

    Student name:
    PhD
    Other supervisor: Dr Stephen Lindsay
    Other supervisor: Prof Gareth Stratton