This is the home for open code, data and research publications for MLGIG (Machine Learning Research Group of Dr Georgiana Ifrim). Below is a list of our public repositories organised mainly into Data Challenges, Time Series Classification (algorithms, applications), Time Series Explanation, Sequence Learning and Text Mining. The Keynotes section provides overviews of our work over time.
- 2026 XKDD, ECML-PKDD: Evaluating Explanation Methods for Time Series Classification: Challenges and Resources
- 2025 tempXAI, ECML-PKDD: Beyond Accuracy: The Dual Challenge of Effective and Explainable Time Series Classification
- 2024 MLGIG Overview Presentation Provides a short overview of some of our ongoing projects.
- 2023 Text2Story, ECIR23: Structured Summarisation of News at Scale
- 2023 HCAIEP, ML-Labs CRT Overview: Reflections on Using Robot Racing Cars for Teaching Machine Learning
- 2021 AI4Narratives, IJCAI: The Ants Have Megaphones Now: Text Mining and Summarization for News and Social Media Streams
- 2021 ROAIDays: Time Series Classification: Tradeoffs between Accuracy, Efficiency and Explanation
- 2019 NFMCP, ECML-PKDD: Effective Linear Models for Learning with Sequences and Time Series
Where we apply our knowledge of machine learning modeling and time series modeling:
- 2024 VistaMilk-challenge-lactose-prediction-regression (1st place)
- 2022 VistaMilk-challenge-diet-classification (1st place)
- 2021 VistaMilk-spectroscopy-challenge-predict-milk-traits-regression (2nd place)
- 2014 SNOW Data Challenge (1st prize): Twitter Topics (1st place)
- 2026 Random Initialization as a Baseline for Time Series Foundation Models
- 2026 Revisiting WEASEL 2.0: Reproduction, Sensitivity, and an Adaptive Ensemble-Size Rule
- 2025 An Empirical Evaluation of Foundation Models for Multivariate Time Series Classification
- 2024 Back to Basics: A Sanity Check on Modern Time Series Classification Algorithms
- 2023 Fast Channel Selection for Multivariate Time Series Classification
- 2022 MRSQM: Multiple Representation Sequence Miner
- 2020 An Examination of the State-of-the-Art for Multivariate Time Series Classification
- 2019 MRSEQL: Multiple Representation Sequence Learner
- 2026 Beyond Simulated Benchmarks: Evaluating Motion Representations for Fall Detection Under Real-World Data Scarcity
- 2026 Costream: Cost-Sensitive Streaming Event Detection
- 2024 Accurate and Efficient Real-World Fall Detection Using Time Series Techniques
- 2023 An Examination of Wearable Sensors versus Video Data Capture for Human Exercise Classification
- 2022 Fast and Robust Video-Based Exercise Classification via Body Pose Tracking and Scalable Multivariate Time Series Classifiers
- 2021 Interpretable Classification of Human Exercise Videos through Pose Estimation and Multivariate Time Series Analysis
- 2026 eHYDRA: An Ante-hoc Explanation Method for the Efficient HYDRA Time Series Classifier
- 2026 drXAI: Scaling Time Series Classification via XAI-Driven Data Reduction
- 2025 TSHAP: Fast and Exact SHAP for Explaining Time Series Classification and Regression
- 2025 An Empirical Evaluation of Factors Affecting SHAP Explanation of Time Series Classification
- 2024 tsCaptum - A library for univariate and multivariate time series explanation
- 2024 Improving the Evaluation and Actionability of Explanation Methods for Multivariate Time Series Classification
- 2024 AMEE: Robust explainer recommendation for time series classification
- 2023 Evaluating-Explanation-Methods-for-MTSC