0
$\begingroup$

I tried using parallel coordinates to visualize outliers. Is it fundamentally correct?

Parallel coordinates visualizing outliers

$\endgroup$

1 Answer 1

1
$\begingroup$

Scatter plot and box plots are the most preferred for visualizing outliers.

Parallel plots can also be utilized for detecting outliers. For large datasets it can be bit confusing, highlighting outliers comes in handy then.

$\endgroup$
2
  • $\begingroup$ My data is multi-dimensional with feature values ranging from tens to thousands and I want to visualize whole instances so I can't scatter or box plot. Is parallel coordinates the best choice here? $\endgroup$ Jul 23, 2021 at 6:34
  • 1
    $\begingroup$ Yes, as mentioned in my answer for large datasets that's the typical choice. You can look into the links I have already provided. $\endgroup$ Jul 23, 2021 at 7:01

Not the answer you're looking for? Browse other questions tagged or ask your own question.