Project: AUD-it: Visual Exploration of Intervals in Multivariate Event Sequences
Description
The recent COVID-19 pandemic introduced new challenges for mental healthcare, such as the detrimental effects it has on peoples mental well-being, but also the limited access that people have to mental healthcare systems.
PROSIT (Predicting Risk and Outcomes of Social Interactions) is a project that explores if mental health can be monitored by using an app that tracks regular phone data. Currently, this data is audited by researchers from PROSIT in a spreadsheet-like format. This is time consuming and prone to errors. In this format, it is difficult to explore and validate event-based data in a temporal context, and in context to other events.
We propose AUD-it, a visualization tool that overcomes these problems through interactive visualizations of event-based data. This tool addresses research gaps that relate to mobile health visualizations and event sequence visualizations. We do this in terms of evaluation and optimization from a research-based perspective, integrating contextual information about relations between events and temporal aspects, and proposing atypical combinations of interactions and visualizations to improve aggregate analyses.
Moreover, we show that this tool can be used for a dataset which contains low-level events that have been recorded in uncontrolled environments. AUD-it has been tested with four researchers from PROSIT. The researchers were able to identify and solve use cases with AUD-it, and they evaluate the system as an improvement over the current method they use to analyze the data.
Details
- Student
-
EFEvie de Fouw
- Supervisor
-
Anna Vilanova
- Secondary supervisor
-
FLFernando Paulovich, Sanne van der Linden
- Link
- Thesis