What is the meaning of jitter in the weka tab render

Question

In weka, I upload an arff file. I can view the relationship between attributes using the visualization tab.

However, I can not understand the meaning of the jitter slider. What is his purpose?

Answer 1

You can find the answer in the mailing list archives :

The jitter feature on the visualization panel simply adds artificial random noise to the coordinates of the plotted points to spread the data a bit (so that you can see points that might be hidden by others). A.

Answer 2

I do not know weka, but as a rule, jitter is the term for changing a periodic signal to some reference interval. I assume that the slider allows you to set some range or threshold below which data points are considered regular, or change the output to make some changes. A wikipedia entry may give you some background.

Update: from this pdf , the jitter slider is used for this purpose:

Jitter for handling nominal attributes (and for detecting "hidden" data points)

Based on the accompanying slide, it looks like it is introducing some changes in the visualization, possibly to show when two data points overlap.

Update 2: This google books compilation (for data mining by Jan Witten, Abe Frank) seems to confirm my assumption:

[jitter] - a random offset applied to the X and Y values ​​to separate points lying one above the other. Without jitter, 1000 instances at the same data point look exactly the same as 1 instance

Answer 3

I don’t know the products you mentioned, but jittering usually means randomizing sample positions. For example, when ray tracing, you usually render the ray, although every pixel on the screen. Jittering adds a random bias to each beam to reduce problems caused by a regular alias.