From the course: Data Science Foundations: Data Assessment for Predictive Modeling
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
How to navigate borderline cases of variable type
From the course: Data Science Foundations: Data Assessment for Predictive Modeling
How to navigate borderline cases of variable type
- [Instructor] Okay, we're in KNIME with the same dataset, and the same workflow. So now what we're going to talk about is variables that seem like they might be one level of measurement but where there's some benefit to call them another level of measurement. I'll illustrate with an example. We're going to use hours. So the way that we're supposed to look at hours is with a histogram. So I start to types that in. And there's an interactive histogram, which we can use. We'll hook that up and when we go in, we don't want final weight, we want hours-per-week. And we definitely want to display all rows. Click on OK. Execute and open views. Okay, so we can see a pattern here. We can see that the tallest bar up here appears to be 33 to 41. Although 25 to 33 is a pretty tall bar as well. The problem is we can't see who said exactly 40. And just instinctively, we know that exactly 40 is going to be a common choice.…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Contents
-
-
-
-
-
-
-
-
-
(Locked)
The explore data task1m 1s
-
(Locked)
How to be effective doing univariate analysis and data visualization3m 18s
-
(Locked)
Anscombe's quartet6m 26s
-
(Locked)
The Data Explorer node feature in KNIME5m 14s
-
(Locked)
How to navigate borderline cases of variable type5m 11s
-
(Locked)
How to be effective in doing bivariate data visualization8m 34s
-
(Locked)
Challenge: Producing bivariate visualizations for case study 11m 18s
-
(Locked)
Solution: Producing bivariate visualizations for case study 15m 40s
-
(Locked)
-
-
-
-
-
-