The AMIS Secretariat organized a research seminar dedicated to the work on market predictions using sentiment derived from news articles. The event was held back-to-back with the 12th Session of the Global Food Market Information Group at the headquarters of the World Trade Organization in Geneva.

The research on sentiment-based prediction is led by the Food and Agriculture Organization of the United Nations (FAO). The project team presented an overview of the concepts of artificial intelligence, machine learning and deep learning, which are at the base of the sentiment work, before providing more information on the prediction mechanism itself. Following this presentation, GEOGLAM gave an overview of machine learning in crop forecasting using remotely-sensed data.

The rich discussion that followed demonstrated the potential and large interest in applications such as sentiment-based market predictions. FAO was encouraged to finalise the development and testing of the model and conceive a strategy for dissemination.