We are currently advertising DTP and CASE studentships for October 2019 start. The deadline for applications is the 26th November 2018.

26 / 09 / 2018

Predicting and understanding crop yield with machine learning (BAGNALL_U19DTP)

how to apply

Accurately predicting the yield of core crops such as potatoes or wheat is central to addressing the challenges of food security and reducing the impact of climate change [1].  Current techniques for predictions are often inaccurate and rely on human expertise. This project will use a range of machine learning algorithms to form predictions of crop yield under two scenarios. The first will use data from sources such as google earth and the met office to build models over the whole of the UK. The second approach will use an internet of things sensor net approach to construct localised models for much smaller regions. The successful candidate will join a vibrant machine learning research group working on a range of real world problems and will have the opportunity to collaborate with researchers on three continents.  We will experiment with deep learning solutions and a range of alternative machine learning techniques developed at UEA [2,3]. The project will suit a proficient programmer with some exposure to machine learning. 


[1] J. Lou et al.  (2017) Deep Gaussian Process for Crop Yield Prediction Based on Remote Sensing Data in Proc AAAI.

[2] Lines et al. (2018). Time Series Classification with HIVE-COTE: The Hierarchical Vote Collective of Transformation-based Ensembles. ACM Transactions on Knowledge Discovery from Data 12(5)

[3] Bagnall et al. (2017). The great time series classification bake off: a review and experimental evaluation of recent algorithmic advances, Data Mining and Knowledge Discovery, 31(3).