Duration: 60 minutes
Track: Analyze & Interpret
Every predictive modeler or analytics professional knows the importance of data preparation; experts place the time expended in this stage from 50% to even 90% of the time one spends building predictive models. It is time-consuming and daunting because there are so many ways data can be wrong. Yet there are many principles that are reused in nearly every data set.
This session focuses on three essential steps in data preparation, taking into consideration the data itself and how algorithms sometimes dictate the kinds of data preparation we do. The principles apply not only to predictive models, but also to data visualization and dashboards. Examples from actual modeling projects will illustrate the principles.
To access the session recording, please login or register first:
Not a PASS member? Click here to register.
Dean Abbott is Co-Founder and Chief Data Scientist of SmarterHQ. Mr. Abbott is an internationally recognized data mining and predictive analytics expert with three decades of experience applying... View More