In the last post by Vedant Bahel you learnt about what is Supervised Learning. In this post we will dive deeper into supervised learning and discuss the subcategories in which supervised machine learning problems can be divided.
Supervised Learning
So, basically every supervised learning problem can be divided into these two categories
Regression Problem
In a regression problem, we try to predict results within a continuous output, meaning that we are trying to map input variables to some continuous function.
Like predicting the ages people, salaries of employees etc. Something that can take values in a wide range and isn’t quite easy to predict.
A classic example could be prediction of housing prices given the size of the house.
As we see here the output varies according to the input and isn’t confined to any particular range.
Classification Problem
In a classification problem, we try to predict results in a discrete output. In other words, we are trying to map input variables into discrete categories.
It is like classifying:
- Whether a tumor is Benign or Malignant (2 Categories)
- Whether an object is red or blue or some other colored (3 Categories)
- Whether an Email is spam or not a spam (2 Categories)
Here we are trying to classify our output into one of the two categories or more categories present. Thus we are sure that all our input will lie in one of these categories. Like we know that the tumor will be either benign or malignant, it can’t be anything else.
Whereas in a regression problem our output can lie anywhere depending on the input.
That’s all for today! We’ll learn more about Regression Problems more in the next post. So stay connected!
Tech Eagle 