-1
$\begingroup$

I am doing regression analysis on a data set with over 20000 samples using scikit learn. Trying to use regression models to fit three features to label which is a score ranges from 0 to 10. Problem is only 100 of the data has a known score. The rest are all unlabeled.

Semi-supervised learning seems to work well with classification problems using methods like label-propagation. I wonder if it works for regression problems as well. If so, where can I find any examples for labeling unlabeled data based on similarity.

$\endgroup$
  • $\begingroup$ If the scores are only integers from 0 to 10, did you try semi-supervised classification on this data? $\endgroup$ – rnso Nov 4 '18 at 2:10
0
$\begingroup$

I also had a similar dataset, I came across covarite shift technique of Machine Learning. As you have changes in the distribution of the input variables in training data. Though the predictions might also be wrong using this.

| improve this answer | |
$\endgroup$

Your Answer

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

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