This project, carried out in conjunction with a major global automotive manufacturer, aims to develop and validate machine learning algorithms for semantic processing of heterogeneous data available from operational automotive systems warranty databases, and the automatic generation of data models. This research will consider challenges with machine learning in multidimensional heterogeneous data sets (both text and variable data) stemming from the multiple types of data, accuracy and imbalance in the real world operational reliability data sets, which affects the robustness of the machine learning models. The validation of the methods and algorithms developed will be carried out in conjunction with real world data available from the automotive Company partner.
The ideal candidate will have a good background in machine learning and data analytics applied to an engineering context.
The project will be based in the interdisciplinary Advanced Automotive Analytics Research Laboratory, part of the University of Bradford Automotive Research Centre, which has a strong track record of collaborations with the global automotive industry spanning over 25 years. The project is expected to start no later than June 2022.