AI and Ensemble approaches to model transient lipid deterioration in particulate food systems

脂类是食品的主要成分，价值约500亿美元。不饱和脂质成分（如亚油酸、花生四烯酸、二十碳五烯酸和二十二碳六烯酸）会被氧化，从而导致风味和口感恶化，并降低营养价值，从而严重影响食品品质和货架稳定性。该项目假设人工智能和机器学习可以预测产品中发生的质量变化，或者制定稳定的新食品。这种模式大大减少了食物浪费、时间和资源。学生将在雷丁大学获得大量关于货架寿命和保持质量评估的实验和数学建模方法的培训。此外，学生将接受使用最新机器学习和人工智能方法的培训，并将其应用于食品系统-这仍然是一个处于起步阶段的领域。

Lipids are a major constituent of foods which are worth around USD 50 Billion. Unsaturated lipid components [e.g. linoleic, arachidonic, eicosapentaenoic and docosahexaenoic acids] undergo oxidation which severely impacts on food quality and shelf-stability through flavour and taste deterioration and decreases in nutritive value. This project hypothesises that Artificial Intelligence and Machine Learning allow prediction of quality changes occurring in a product, or to formulate new food products which are stable. Such models result in a substantial reduction in food waste, time and resources.The student will gain considerable training in experimental and mathematical modelling methods used in shelf-life and keeping quality assessment at the University of Reading. In addition, the student will be trained in the use of latest Machine Learning and Artificial Intelligence methods and apply these to food systems - which is still an area in its infancy.