Motivation:

Often the first step in analyzing a dataset is thinking of different ways of visualizing your raw data: exploratory data visualizations. In general these visualizations are only used by the data scientist for personal use. After deriving some insights these insights can be communicated using explanatorydata visualizations. There is actually a problem with this approach: Data scientist are generally not good at all aspects of Data Science. Some people are experts in Machine Learning, some are code gurus but in general the third collection (figure on the right), being the domain expertise is somewhat ignored. Can one derive insights from data without extensive domain knowledge?

In this context the transformation from exploratory to explanatory data analysis is problematic since this transformation is performed by the data scientist and during this transformation information gets lost.

Therefore a third approach is feasible. Can exploratory data visualizations be an…