Can we find a proportion that maximize the distance of the SVM?

[NeighborhoodCoding]

I think we can find the proportion that maximize the distance of the SVM(= the distance of hyperplane)

for e.g : contamination proportion 0.1% --> distance of SVM is 0.1(normalized for data range, 0 to 1) contamination proportion 0.5% --> distance of SVM is 1(normalized for data range, 0 to 1) contamination proportion 1% --> distance of SVM is 0.1(normalized for data range, 0 to 1)

and we can guess, this input data may have the outliers about 0.5% and if we use the contamination proportion to "auto", pyod can find the proportion of the outliers like above process.

However, I'm beginner of coding and theory. I think, finding the contamination proportion that maximize the distance of the SVM may have some meaning. thx.