Back

Explore Courses Blog Tutorials Interview Questions
0 votes
2 views
in Machine Learning by (55.6k points)

Can anyone explain overfitting and what to do to overcome overfitting?

2 Answers

0 votes
by (119k points)

Overfitting in Machine learning is making our model learn too much from the training dataset. If we train too much on the training dataset, it learns the noise from the dataset also. So, when we build this model on unseen data then it fails to predict the results accurately. This can be solved by increasing the training data or by using regularization or ensemble techniques.

If you want to learn Machine learning, you can enroll this Machine learning online course by Intellipaat.

You can watch this video on machine learning by Intellipaat before starting to learn machine learning:

0 votes
by (108k points)

Overfitting is a condition in Machine Learning where a model exhausts its ability to find effective results by overanalyzing the input. Overanalyzing comes from the fact that the model dwells so deep within data that it finds and uses noise rather than useful chunks of the data to learn. This causes model instability and lower efficiency. Overfitting happens with models that are nonparametric and nonlinear, usually because of the extra headroom they have in terms of flexibility to learn. Decision trees are sometimes subjected to overfitting as they are a type of nonparametric algorithm with high flexibility. However, this can be quickly fixed by pruning the tree after the process of learning to curb some about of the details.

If you are looking for an online course to learn Machine Learning, I recommend this Machine Learning Certification program by Intellipaat.

Browse Categories

...