Although visual depictions are easier to understand than other forms of data representation, users still need to learn how to read and understand them. While journalists or biomedical specialists may be experts in their respective fields, most have difficulty interpreting and working with new visual representations or with understanding methods of data analytics. This harbours the risk of drawing false conclusions as well as triggering frustration or rejection of powerful data tools. Since VA is a young and dynamic discipline, a wide range of new VA approaches are constantly emerging.
Onboarding methods aim to help users to understand data visualisation and automatic analysis algorithms and to fully exploit the possibilities offered by the tools available – a topic previously neglected by science. Previous solutions have concentrated primarily on the user interface and not on visual representations. Furthermore, such assistance systems must be custom built, which involves considerable effort and thus places a financial burden on SMEs in particular. Empowering users through VA onboarding methods is an open research task and is crucial for providers of software solutions. The company partners within the consortium produce VA tools for biomedical specialists and data journalists. For them, the rapid applicability of visual data tools is one of the core challenges on the market. The possibility of using this technology in their products would be a significant market advantage, since the new tools increase user friendliness and improve the quality of results at a moderate cost.
The aim of the project is to develop suitable onboarding methods for VA tools which can be automatically generated. Our vision is to have self-explanatory tools which actively support users in interpreting visualisations and analysis methods. The results will include a conceptual framework for VA onboarding, proof-of-concept implementations and applications in biomedicine and journalism.