Lecturer, University of East Anglia
Jason Lines is a Lecturer in the School of Computing Sciences at the University of East Anglia (UEA) in Norwich. He previously worked as a Senior Research Associate at the UEA and is also affiliated with the Alan Turing Institute having been seconded on a Project Fellowship at the end of 2018.
Jason’s research interests are in the fields of machine learning and artificial intelligence, with a specific focus on classification and time series data. His recent research has investigated the use of heterogenous ensembles across various data representations for time series classification (TSC), leading to the current state-of-the-art algorithm for TSC (HIVE-COTE). He has also recently collaborated on one of the largest ever machine learning studies, The Great Time Series Classification Bake Off, which involved over 30 million individual experiments to benchmark the best-in-class algorithms for TSC; this work was formally published at the end of 2017 and was among Springer’s top-10 downloaded Computer Science papers in 2018, with over 21 thousand downloads on Springer’s website as of early 2019.
Time Series Classification with HIVE-COTE: The Hierarchical Vote Collective of Transformation-based Ensembles Lines, J., Taylor, S. & Bagnall, A. Jul 2018 In ACM Transactions on Knowledge Discovery from Data. 12, 5, 35 p., 52
The Great Time Series Classification Bake Off: a Review and Experimental Evaluation of Recent Algorithmic Advances Bagnall, A., Lines, J., Bostrom, A., Large, J. & Keogh, E. May 2017 In Data Mining and Knowledge Discovery. 31, 3, p. 606–660
Time-Series Classification with COTE: The Collective of Transformation-Based Ensembles Bagnall, A., Lines, J., Hills, J. & Bostrom, A. 1 Sep 2015 In IEEE Transactions on Knowledge and Data Engineering. 27, 9, p. 2522-2535 14 p.
Lines, J. & Bagnall, A. May 2015 In Data Mining and Knowledge Discovery. 29, 3, p. 565-592 28 p.