-2

those're the learning curves for each algorithm I used. i'm working on my report and i'm confused how to interpret the curve. I used multi label classification algorithms. this is the learning curve of binary relevance the classifier is KNeighborsClassifier. 1st learning curve

the second one is the curve of classifier chain using DecisionTreeClassifier 2nd learning curve

and the last one is the curve of LabelPowerset using GaussianNB 3rd learning curve

which one is the best? because the accuracy and the F1 score are good results

najoukou
  • 27
  • 5
  • It's suspicious that the second one already gives you 100% Training score with few samples. You might be overfitting. How are the three methods performing in the test set? –  Jun 01 '22 at 08:55
  • yeah true the dataset in not big. i got good result in the classification report. i think the second one is overfitting how can i proveit – najoukou Jun 01 '22 at 09:44

1 Answers1

0

Learning curves are a tool to finding which models benefit from increasing training data. In other words, they indicate whether a model, with an increased dataset, will give better results.

The best curve in my opinion is the one that gives the best normalized score in a minimum training example. It also has to converge fast enough to a good score.

Maxime Bouhadana
  • 438
  • 2
  • 10