This document aims at describing the Multimodal Forecasting and Causal Techniques module prototype which will form a part of the health analytics layer and will be used to perform several analytical procedures that will aid the policy making process. The initial prototype reflects the work that has been done in the scope of task T5.6 (Multimodal Forecasting and Causal Techniques) between M04-M11 and describes the main components of the prototype as well as the basic technologies used in its development.
Due to limited availability of data from the UCs, the current prototype is based exclusively on the SLOfit use case (ULJ and JSI). The Multimodal Forecasting and Causal Techniques module will eventually consist of two main building blocks: i) Causal analysis, ii) Forecasting. However, at this time only the forecasting module is included in the prototype. As a result, this document not only describes the currently available functionalities, but also gives a high level overview of the future components to be implemented in the next version of the prototype. Both causal analysis and forecasting of individual parameters will represent progress beyond the current state-of-the-art.
As the work to be done in the scope of T5.6 is not yet finalised, updated versions of this document are planned to be delivered during the process of the project (M22, M34) which will include the additional components and functionalities that will be implemented.