I have a group of element. Every element have 5 criteria. I want to know if a specific element can be representative of the group, I am looking for the best percentage of resemblance.
I have a group of element. Let say 5 elements, each one of them have a certain number of criteria (5 for example). I want to know if it is possible to say that element 1 for example represents the other elements in term of criteria, for example by 30% for first criteria, 20% for second criteria and so on.
I did try to look for the mean value of each criteria for the 5 elements and then I looked for distance between each element and the mean and I select the element that have the shorter distance to the mean as the most representative of the group but I didn’t know if it is correct or not, and I prefer not to use one mean and one criteria but a mix of all them. I hope that my explanation is good
You can, for example, annotate the 5 elements according to your criteria, cluster the elements using, for example, k-means, and look how centers of clusters are remoted from the elements.