I am relatively new to KNIME and are trying to do similarity searches, between a dataset containing fingerprints and a dataset containing reference fingerprints.
However, for the similarity search I want to make use of less common similarity coefficients that are not incorporated within, for instance, the similarity search node:
Simple matching: s= (a+d)/(a+b+c+d)
Sokal-Sneath: s= 1/4 [a/(a+b)+a/(a+c)+d/(b+d)+d/(c+d)]
In which a, b, c and d are:
- a = number of bits set in both fingerprints (=1)
- d = number of bits not set in both fingerprints (=0)
- b = number of bits set in fingerprint but not in reference fingerprint
- c = number of bits not set in fingerprint but set in reference fingerprint
I thought that the use of the similarity search node with a user provided distance measure (using the Java distance node) would be the solution. However, I cannot figure out how to make this work.
Any help or suggestions would be highly appreciated.