FindAPhD Weekly PhD Newsletter | JOIN NOW FindAPhD Weekly PhD Newsletter | JOIN NOW

Anomaly Detection in Time Series


   School of Computing Sciences

This project is no longer listed on FindAPhD.com and may not be available.

Click here to search FindAPhD.com for PhD studentship opportunities
  Prof T Bagnall  No more applications being accepted  Self-Funded PhD Students Only

About the Project

Time series classification (TSC) involves building predictive models for problems where the attributes are ordered. So, for example, we may want to predict whether a patient is healthy or not based on their heartbeat. UEA has been at the forefront in developing algorithms for this type of problem [1,2,3]. TSC assumes that each observation is independent (e.g. a separate patient). An alternative common scenario is when we have a time series where readings are continuously taken, and we wish to detect when a specific event occurs. We may, for example, want to detect when a person falls over based on a motion sensor that is continuously collecting data. There are a range of techniques for doing this that are closely related to classification algorithms. The student will work in a vibrant research group to design algorithms within open source code frameworks [4] and evaluate them on a range of real world problems. There will be opportunities to work with our collaborators across the world, including researchers at the Alan Turing Institute, the University of California, USA and Monash University, Australia. 


Funding Notes

This PhD project is offered on a self-funding basis. It is open to applicants with funding or those applying to funding sources. Details of tuition fees can be found at https://www.uea.ac.uk/about/university-information/finance-and-procurement/finance-information-for-students/tuition-fees
A bench fee is also payable on top of the tuition fee to cover specialist equipment or laboratory costs required for the research. Applicants should contact the primary supervisor for further information about the fee associated with the project.

References

[1] Gharghabi, S., Imani, S., Bagnall, A., Darvishzadeh, A. and Keogh, E., An Ultra-Fast Time Series Distance Measure to allow Data Mining in more Complex Real-World Deployments, Data Mining and Knowledge Discovery. 34(4), 1104-1135, 2020.
https://link.springer.com/article/10.1007/s10618-018-0565-y
[2] Lines, J., Taylor, S. and Bagnall, A Time Series Classification with HIVE-COTE: The Hierarchical Vote Collective of Transformation-based Ensembles. ACM Transactions on Knowledge Discovery from Data. 12(5): 52-87, 2018.
https://dl.acm.org/citation.cfm?id=3182382
[3] Bagnall, A., Lines, J., Bostrom, A., Large, J. and Keogh, E. The Great Time Series Classification Bake Off: a Review and Experimental Evaluation of Recent Algorithmic Advances. Data Mining and Knowledge Discovery, 31(3): 606-660, 2017
https://link.springer.com/article/10.1007/s10618-016-0483-9
[4] Löning, M. Bagnall, A., Ganesh, S., Kazakov, V., Lines, J and Király, F.. sktime: A Unified Interface for Machine Learning with Time Series, Workshop on ML for Systems at NeurIPS, 2019
https://github.com/alan-turing-institute/sktime
[5] Middlehurst, M., Large, J., Flynn, M. et al. HIVE-COTE 2.0: a new meta ensemble for time series classification. Machine Learning (2021).
https://doi.org/10.1007/s10994-021-06057-9
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs