Supervised Learning - As the name suggest, Supervised Learning is a process of learning with the help of a instructor who explain complex topics in simple ways. We can learn all about the topic in much easier ways and can excel in it.
In Machine Learning, we train a machine or an algorithm through labeled data and featured data in the form of input - output pairs. Based on these pairs, It calculates a prediction. Now we can test the algorithm by providing a new input and it should give us the desired result.
We repeatedly train and test data until the algorithm achieves an acceptable outcome. Machine or algorithm is working on the basis of initial input-output pairs we provided to it.
Steps for Supervised Machine Learning:
We will gather all the desired information.
We will feed all this information with the help of some methods to train our machine.
We test our machine by giving new input.
Example: In above table, we provided our machine an input - output pairs of four houses having different values, where first three columns are input and Rent is the output. The algorithm will recongize the patterns between market rates, carpet area, bathrooms and bedrooms.
Based on above input-output pairs, it should predict the rent of a desired house whose Carpet area is 900, 2 bedrooms and 1 bathroom as 1250.
Supervised Learning can be divided into various types, the most popular ones are:
Classification: A classification problem is when the output variable is a categorical data, meaning the answer to our prediction can only be the one of the specified categories.
Some examples are:
If the color is Red or Blue
This patient has a disease or not
2. Regression: When the output variable is a continuous value, meaning it can be anything.
Like in above example, we can not specify the range of rent. It can be anything
Predicting tomorrow’s temperature, it can be anything but if we look for a specific answer like is it going to rain, then it becomes a Classification problem
Benefits of Supervised Learning
In supervised learning, we can easily understand the process.
We know all the features of our input.
We can set the specific boundaries of our features.