BigDataGrapes D5.2 - Uncertainty-aware Visual Analytic Components

In this deliverable (D5.2), we demonstrate uncertainty-aware visual analytic components that highlight uncertainties in prediction outcomes. Previous research has shown the importance of showing uncertainty as they could improve awareness and trust of the readers, particularly nontechnical experts. With a growing awareness of requirements to convey uncertainty to non-expert users, researchers have proposed to extend traditional visualisation techniques to represent the uncertainty information associated with data. Nonetheless, the application of uncertainty visualisations is limited in the agriculture domain.

To demonstrate our components, in the previous version of this deliverable, we used a dataset from one of our pilot partners, INRA. The dataset contained alcoholic fermentation kinetics measured over time. The main attributes present in the dataset were the level of carbon dioxide (CO2), temperature and time at measure. We demonstrated the possibilities of various uncertainty representation techniques.

In the current version of deliverable 5.2, we have refined the visualisations and updated the dataset to be in line with end-user evaluation requirements. Specifically, we use the price prediction scenario of agricultural products, proposed by Agroknow, to demonstrate the updated uncertainty visualisations. End-user evaluation, as part of WP8, is planned for July 2020 where our updated version of uncertainty-aware visual analytic components, along with others, will be evaluated by potential users from the industry. In this document, we present our updated version of uncertainty-aware visual analytic components and the development framework used to implement the components.

This document is structured as follows. Chapter 1 lays out an introduction to the deliverable describing the existing work and motivations. In Chapter 2, the visualisation components together with their development framework are described in detail. This document concludes with Chapter 3 where a summary of the deliverable is underlined.