Building analytic datasets
Analytical datasets are semi-denormalized tables. By semi-denormalized, this means including not just the ID code of a field but the description for it as well. You may also decide to create categories based on value ranges and include these as separate features.
The goal is to make life easy for your analysts more than a focus on efficiently storing values, as it would be with purely relational database design. You are, essentially, prebuilding the transformed datasets that an analyst would be building using SQL to preprocess a dataset in preparation to train an ML model anyway.
As data wrangling takes 80-95% of a typical analysts time, having most of this already built, tested, and the business logic incorporated ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access