2

I want to compare two sets of keywords. After searching I understood that Jiang-Conrath Similarity is closer correlating with human judgement. I read this definition:

Jiang-Conrath Similarity Return a score denoting how similar two word senses are, based on the Information Content (IC) of the Least Common Subsumer (most specific ancestor node) and that of the two input Synsets. The relationship is given by the equation 1 / (IC(s1) + IC(s2) - 2 * IC(lcs)).

my question is that how I can choose a threshold to say that two words are simililar with Jiang-Conrath measurement. For example, similarity of cat and dog is only 0.4 with Brown-ic.

Improve this question
1
  • 1
    I suggest some experimentation; try some word pairs you consider more or less similar and compute the JC similarity for them and compare with your judgements.
    – Sir Cornflakes
    Commented Jan 4, 2016 at 16:40

1 Answer 1

Reset to default
3

I think the result you can find out there would be the experimentation of someone else. So my suggestion is make your own experimentation: build your own classifier.

A good and simple idea

  • collect as many classified data as you can
  • choose the similarity as a feature and peek the labels as similar and non-similar

And, assuming the data is linearly separable, find the farthest value from the two labelled sets using SVM (Support Vector Machine) with a linear kernel.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.