Minimize Machine learning for yield forecasting using EO and meteorological data

Date added: 21 November 2019

Organisation: JRC

Location: Ispra, Italy

Description:

The Food Security Unit (European Commission - Joint Research Centre, Ispra, Italy) is opening a Contract Agent position on machine learning and deep learning for yield forecasting in Africa using Earth observation and meteorological data.

The successful candidate will develop yield forecasting methods using ML and compare them, in terms of accuracy and timeliness, against the currently used ones (mostly multiple regressions using remote sensing and meteorological data). The position is for one initial year (renewable up to six) and open for citizens of EU member states and associate countries

 

Role Tasks:

The successful candidate will be in charge of:

  • Designing, setting up, and coding ML methods (including random forest, support vector machine, neural network and 1D convolutional neural network) to forecast crop yield in selected African countries;
  • Testing the performances of ML methods, in terms of both accuracy and timeliness of the prediction, against benchmarking results obtained by the CST;
  • Dissemination/publication of the results.
  • contributing to partnerships and networks with African partners and, according to needs, support capacity building and training.
  • Qualifications

The ideal candidate should have:

  • Completed university studies of at least three years attested by a diploma and at least three years professional
  • experience in a field relevant to the position, or a doctoral diploma in geospatial sciences, remote sensing, image processing, machine learning, big data analytics, applied mathematics and computer science, bioscience and agriculture or related field;
  • Excellent programming skills in python are essential, C/C++ or other languages are advantage;
  • Experience with deep learning and ML packages (e.g. TensorFlow, Keras, PyTorch) are essential, experience with explainable artificial intelligence is an asset;
  • Experience in application of ML to regression problems in geospatial data is an advantage;
  • Solid record.

Learn more about the opportunity and how to apply.


Minimize Useful Links
eoPortal News Missions Directory Earth Observation Images Earth Observation Events 
Earth Observation Jobs about eoportal Contact eoPortal Follow us on Twitter