Matthew Middlehurst
Lecturer In Computer Science at University of Bradford
Bradford, England, United Kingdom
Summary
🤩
Rockstar🎓
Top SchoolMatthew Middlehurst is a Lecturer in Computer Science at the University of Bradford with seven years' experience developing and teaching machine learning for time series. He researches time series ML algorithms and is a core developer for aeon, contributing production-ready implementations and refactors that bridge novel research and usable open-source tooling. His work on sktime and aeon shows a pragmatic focus on improving transformer interfaces, classifier performance, and reproducible pipelines that help researchers adopt new methods quickly. Matthew combines hands-on software engineering with academic rigour—teaching programming and ML across undergraduate and master's courses while maintaining widely used libraries. Colleagues rely on him for translating experimental ideas into robust code, and he has a track record of integrating external innovations (e.g., SFA and Weasel features) into unified frameworks.
7 years of coding experience
5 years of employment as a software developer
BTEC Level 3 Extended Diploma in Computing, D*D*D*, BTEC Level 3 Extended Diploma in Computing, D*D*D* at Cambridge Regional CollegeDoctor of Philosophy - PhD, Computer Science, Doctor of Philosophy - PhD, Computer Science at University of East Anglia
Github Skills (11)
scikit-learn10
data-analysis10
machine-learning10
ensembles10
python10
scikit10
classify9
classification9
time-series9
data-mining9
code-optimization8
Programming languages (8)
Github contributions (5)
aeon-toolkit/aeon
Feb 2023 - Mar 2023
A toolkit for machine learning from time series
Role in this project:
ML Engineer
ML EngineerContributions:23 releases, 1251 reviews, 65 commits in 1 month
Contributions summary:Matthew contributed to the core machine-learning functionality of the aeon toolkit by implementing and refactoring methods within the `BaseTransformer` class, splitting `fit` and `transform` functions. They also made bug fixes and enhancements to existing transformers. Additionally, they modified classification-related files and examples, including renaming `convolution_based` to align with the toolkit's architecture and dependencies.
data-miningdata-sciencemachine-learningscikit-learntime-series
sktime/sktime
Jun 2019 - Jan 2023
A unified framework for machine learning with time series
Role in this project:Back-end Developer
Back-end DeveloperContributions:235 reviews, 416 commits, 89 PRs in 3 years 7 months
Contributions summary:Matthew appears to be involved in the development of a Time Series Classifier (TSC) project, as evidenced by the code diffs which feature modifications to existing estimators and transformations. Their work focuses on improving and integrating functions associated with the SFA transform, and making various changes to improve performance. The user has made additions and bug fixes in the direction of integrating new features from other sources such as Weasel.
forecastingtime-series-analysistime-series-regressiondata-sciencedeep-learning
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.