What is difference between unsupervised learning, supervised learning and semi-supervised learning?



While searching about various techniques of machine learning.I have found that there exist three type of learning

Unsupervised Learning

Unsupervised learning is when you have no labeled data available for training. Examples of this are often clustering methods.

Supervised Learning

In this case, your training data exists out of labeled data. The problem you solve here is often predicting the labels for data points without a label.

But I am not able to understand what is semi-supervised learning.



Hi @hinduja1234,

In semi-supervised learning, there exists both labelled and unlabelled data and you have to train your model that learns from both of these.

What you do is first train a model in a supervised way, gaining a basic understanding and then by acquiring more information from clustering (unsupervised learning), you make your model more robust.

This diagram (taken from scikit-learn) shows the output of semi-supervised model.

What is semi-Supervised Learning?