# Performing clustering without a distance matrix

I have n vectors and a matrix of similarity scores between them (e.g. vector 1 score of similarity with vector 4 is 1.3, and with vector 7 is 2.3). This matrix is partial (I don't know what is the similarity between many pairs of vectors) and the similarity is not a distance metric (e.g the triangle inequality doesn't hold). Is there still an efficient way to cluster the vectors such that similar vectors would end up in the same clusters?

Thanks.

• Can you convert the similarity measure into a distance like d(x, y) = 1 - sim(x, y). – Ray Dec 14 '18 at 9:16