Characteristics of Symptom Tracking App Data and its Potential Use for XAI

It is common for people with chronic illnesses to keep records about their health, and this is increasingly being done through symptom tracking apps. These apps allow users to log information about the symptoms they experience, and often incorporate other features such as medication and activity tracking. This creates a rich source of data about the individual's health, but at this stage is not clear how beneficial this data is to users. There is an opportunity for this data to be further utilised to assist users in understanding their health conditions. This paper proposes that this could be done through the application of Explainable AI and presents work-in-progress research. The final goal of this research is to utilise Explainable AI to provide personal insights' into users' self-tracking health-related data. As a first step, this work characterises the features of a large publicly available data-set
$(n_{users}=42,283, n_{reports}$
7,976,223) so that the general features of self-reported symptom tracking data may be better understood.