# What is the name of this similarity distance metric?

def distance_metric(seed, base):
num = 0.0
den = 0.0
num = sum(numpy.minimum(seed,base))
den = sum(numpy.maximum(seed,base))
dist = round(1.0 - 1.0*num/den,4)
return dist


The metric is used to gauge similarity in the context of locality sensitive hashing.

Items within a bucket are kept if their distance is < 0.16.

This is the weighted Jaccard Index.

https://en.wikipedia.org/wiki/Jaccard_index#Weighted_Jaccard_similarity_and_distance

This is different from the regular Jaccard Index (Similarity).