Back to overview

Torch: Season 1, Episode 3

From Witchcraft To Science – How Data Science Is Done

Is data science really a mysterious procedure that only a few magicians understand? As a data scientist, I am often confronted with this myth. But is it actually true? In my Torch talk, I will show you a typical data science procedure, called CRISP-ML, so that you can decide for yourself.


Download Whitepaper

In this torch episode, I will show you:

  • How the CRISP-ML cycle provides a solid structure for any data science project.
  • What business questions you have to ask before even starting the project.
  • Why iteration is at the core of data science.

Hello! My name is Lev Kiwi, and I have been working as a data scientist at Trivadis – Part of Accenture for more than 7 years. Besides helping businesses implement intelligent solutions, I also teach data science and data analysis at the Haute École Arc, Neuchâtel. I am known for always challenging ideas and uncompromisingly searching for the best approach to a data science project. When I am not busy with algorithms and statistical figures, I enjoy horse riding and going on walks with my dog. Did you enjoy my Torch talk? Then let’s get in touch – I would love to pass the torch and discuss the world of data science further!


Recently, my team and I used CRISP-ML to help a consumer goods company make more accurate product demand forecasts. What challenges we encountered and how we solved them? In this takeaway, you will find concrete tips for your own data science project – based on the experiences that we made in this real use case.

Download now

Your contact