Supervised Learning

 Supervised Learning:

In Machine learning types we already discussed about Supervised learning, now we dig in deep into it.

What is Supervised Learning?

Supervised Learning is a type of Machine learning used to learn the models from the labeled training data.It helps us to predict the output for future or unseen data. Model accuracy depends on what you trained to the model.

In general way of speaking, we give input as well as it's output to the model in a large quantity. Model will predict the patterns or relation between i/p and o/p. By using this pattern model will predict the output for unseen data (input).

Understanding the Algorithm of Supervised learning:

Supervised Learning flow:

Examples of Supervised learning:

• Gmail filters a new mail ham or spam based on past information what you have considered.
• Weather predictions
• Chatbots

Types of Supervised learning:

Supervised learning problems are categorized into 

1. Regression 
2. Classification

In a Regression problem we are trying to predict results within a continuous output, means we are trying to map input variables to some continuous function.

Example: 

• House rate prediction
• Find the age of person based on Picture

In a Classification problem, we are trying to discrete the data based on characteristics.

Example:

• Predicting whether the patient have disease or not.

Once, the algorithm trained we test it with test data( that is not part of train set). A well trained algorithm can predict well for new test data. For better results the training data must be Regularized.

Regularization is the technique keep data generic, means the training set neither over fitted nor under fitted.