New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Implement ML-c4.5 #10

Open
niedakh opened this Issue Dec 6, 2014 · 1 comment

Comments

2 participants
@niedakh
Contributor

niedakh commented Dec 6, 2014

A. Clare, R.D. King, Knowledge discovery in multi-label phenotype data, in: Proceedings of the 5th European Conference on PKDD, 2001, pp. 42–53.

Multi-Label C4.5 (ML-C4.5 ) [11] is an adaptation of the well known C4.5 algorithm for multi-label learning by allowing multiple labels in the leaves of the tree. Clare et al. [11] modified the formula for calculating entropy (see Eq. (1)) for solving multi-label problems. The modified entropy sums the entropies for each individual class label. The key property of ML-C4.5 is its computational efficiency:

entropy(E)=−∑Ni=1(p(ci)logp(ci)+q(ci)logq(ci))

where E is the set of examples, p(ci) is the relative frequency of class label c i and q(ci)=1−p(ci).

@ralotaibi

This comment has been minimized.

Show comment
Hide comment
@ralotaibi

ralotaibi Aug 29, 2016

Hi
I did ML-C4.5 implementation in Java. Can I ask what probability you show at leaves? The modification which was proposed by Clare didn't mention the leaves distributions. I guess it is only marginal distribution. For example, if we prune or set the minimum number of examples to 5 , do we use the marginal distribution to predict labels?
Reem

Hi
I did ML-C4.5 implementation in Java. Can I ask what probability you show at leaves? The modification which was proposed by Clare didn't mention the leaves distributions. I guess it is only marginal distribution. For example, if we prune or set the minimum number of examples to 5 , do we use the marginal distribution to predict labels?
Reem

@niedakh niedakh added this to Far future work in Scikit-multilearn Oct 2, 2017

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment