Today I've done deep research on feature engineering using the split-transform-combine strategy with pandas, and was able to comprise the data frame the way I wanted. Next step is to make tableau work. For now, the data set has been prepared for viz on Tableau and online, not for model deployment. I'd still have to transform the data further in order to be able to deploy a clustering algorithm and visualise a timeline. Another four pomodori added to the tally.