Towards Visualization Recommendation – A Semi-Automated Domain-Specific Learning Approach

Information visualization is important in science as it helps scientists in exploring, analysing, and presenting both the obvious and less obvious features of their datasets. However, scientists are not typically visualization experts. It is therefore difficult and time-consuming for them to choose the optimal visualization to convey the desired message. To provide a solution for this problem of visualization selection, we propose a semi-automated, context aware visualization recommendation model. In the model, information will be extracted from data and metadata, the latter providing relevant context. This information will be annotated with suitable domain specific operations (like rank abundance), which will be mapped to the relevant visualizations. We also propose an interactive learning workflow for visualization recommendation that will enrich the model from the knowledge gathered from the interaction with the user. We will use biodiversity research as the application domain to guide the concrete instantiation of our approach and its evaluation.