Workshop: Data Science, Soup to Nuts

We are starting off 2021 with something different: I am going to do a series of workshops on the process of working on a data project, specifically a dataset related to retail transactions. Over a series of workshops we will complete a data science project from start to finish, focusing on data exploration, modelling and communication.

All code for my workshops and talks can be found in this GitHub repo:

https://github.com/kaybenleroll/data_workshops

This workshop is contained in the ws_soupnuts_202101/ folder

This series of workshops is done live, with code being commited and pushed to dedicated GitHub repository that can be found here:

https://github.com/kaybenleroll/dds-soup2nuts

Previous workshops are available on the Dublin Data Science YouTube page at:

https://www.youtube.com/playlist?list=PLBNUi5C90QfZ5t1JrgAtz_LXvCB7Eh5iu

In the fifth workshop, we focus on the concepts behind "Buy-Till-You-Die" models, and discuss both the Pareto/Negative Binomial Model and the Beta-Geometric Negative Binomial model. We discuss the basic theory and how these models are fit based on our transaction level data.

Finally, we will discuss how we might use these models for prediction.

No knowledge of any of these techniques is assumed - part of the job is often trying to learn some existing topic as it is applied to the problem at hand.

As before, this workshop will be recorded and made available online afterwards, and I will use Zoom as our platform of choice.