Types of Machine Learning

Types of Machine Learning

Similar to any other sciences, Machine Learning can be classified based on some parameters. Machine Learning has been primarily categorized as Supervised and Unsupervised Learning. There are other smaller branches as well like Reinforcement learning, and recommender systems, but these are two main branches.

Supervised Learning:

For machine learning, data is a primary requirement. The data is, basically, the reason for the classification of machine learning into Supervised and Unsupervised Learning.

In Supervised Learning, the data contains some parameters, also called as features on which we have to base our predictions and the right answers for those features, and we have to predict the correct output for some other set of data in which these labels are not given.

Consider the example of Housing Price Prediction:

We are given, the data which includes information like Size, Location, age, and furniture type, of houses, and for this data, we are also given the true prices.

Now, we train on this data with an aim that when a set of data like Size, Location, age, and furniture type is given, we can accurately predict the correct price for this new data.

Unsupervised Learning:

In unsupervised learning, the data contains some parameters, also called as features but this time we don't have any correct label. We now have to cluster the data into groups.

For example, the input data may contain, information about online articles, and we have to cluster the data topic wise. We do it by finding some keywords which repetitively occur in these articles. Like fashion may occur repetitively in an article associated with fashion and "stock" may occur repetitively in a business article.

We will first start with Supervised Learning and subsequently go into Unsupervised Learning.