distribution of dissimilarities

Yesterday I managed to get the distribution of dissimilarities into an array object. I was considering using this distribution to find a cut-off point or level of confidence to get rid of users who filled the form but who are creating noise in the definition of clusters. As far as I understand the lower part of this distribution curve is represented by people who are very similar to the rest of the population. The upper part is the contrary.

Leave a Reply