Watching an interesting video on data science trends by a Google data scientist who recommended this data set as good practice, I recorded a video that documents my process working on it from start to finish with the aim of getting a quick, initial understanding of the data as a basis for more comprehensive tasks (report design, advanced analytics, forecasting etc.) following that.
As opposed to a prepared training exercise I have left all real life, unforeseen challenges and how I deal with them in there. The process includes:
- Data preparation
- Cleansing
- Clustering
- Basic visualisations
- Simple AI (Key Influencer visual)
I hope this helps other users but I would also be very interested in comments what you would do differently or what steps you would add.
Here you go, this is the pretty much unedited footage of a 40 min session: