Key Take Away Messages¶
In this Starter Kit we have demonstrated how creative visualisations can already reveal interesting patterns and more elaborate insights in your data, even before any complex algorithm is applied. In particular, we have used bike counter data to illustrate how timeline plots, different types of heatmaps, streamgraph plots and scatterplots can be used bringing various kinds of insights to the surface. We showed how such visualisations could not only be used to explain certain characteristics of the data, such as some nodes having more crossings than others, but also to identify global trends, such as an increase in traffic over the years, and seasonal trends, such as fluctuating popularity within a year. Furthermore, we were able to recognize structural patterns, such as distinct weekday and weekend traffic patterns and by that detect outliers, such as weekend traffic patterns that occur on weekdays. Additionally, we discussed how the selection of colormaps can influence the visibility of given effects.
These insights can help in formulating hypotheses to be validated further or can be used in subsequent analysis steps, such as feature engineering and data-driven modelling.
We thank you for completing this video series and hope to welcome you in another AI Starter Kit tutorial.
Authors: EluciDATA LabPermanent URL