I have suggested a semi-supervised approach for the hierarchical multi-label classification task. I have included the MLSMOTE oversampling technique as a pre-processing step, and then evaluate the impact of the oversampling on the semi-supervised learning Below, the obtained results in terms of Micro-f1 and Macro-f1 scores.

enter image description here enter image description here

These two figures are under the labeling ratio 50% labeled, 50% unlabeled.

My question is, is it possible to say that oversampling is beneficial at the end of co-training, because at the last iteration, we get the same Macro-F1 scores, we obtain higher Micro-F1 scores for the method without oversampling ? the reason could be that the latter is biased towards the majority labels which increases the Micro-f1 scores in detriment of Macro-F1 scores. Thanks for help.

  • $\begingroup$ Since there are no posted answers to your question, you may delete it (see our help center article. However, do not vandalize your questions. Be aware that when you posted content on SE, you gave up the ownership of that content under CC BY-SA 3.0. $\endgroup$ – gung - Reinstate Monica Jul 15 '19 at 15:20

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.