Last Updated on November 6, 2020
Machine learning algorithms are categorized into four parts. Here you will know most two-part of Machine learning that is supervised and unsupervised learning.
- Supervised learning
- Unsupervised learning
- Semi-supervised learning
- Reinforcement learning
People often get confused about supervised learning and unsupervised learning. Here we will discuss supervised and unsupervised learning with example. So, you could understand it easily.
In this task, you trained the machine using the labeled data(defined data) and you know what your output will be. It helps you correct your algorithm if it gives the wrong answer.
To understand more clearly, let’s take an example:-
Did you notice how a child learns new things? Suppose we(Supervisor) put Apple and Banana on the table and told the child about the name of fruits. So, the child kept in his mind that if the color is red and the shape is round, then it is Apple and if its color is yellow and the shape is not round, then it is Banana.
Next time when we put Apple before him, he recognizes it as Apple. How?
Based on prior experience!!!!
This is how supervised learning works.
Applications of supervised learning:-
This is one of the most used applications of our daily lives. If you split it, the word ‘Bio’ and Informatics’, you get the meaning i.e. collecting biological data such as fingerprints, iris, etc. Today our smartphones are capable of learning our biological information for securing our digital information.
2. Speech recognition
This application where you teach the algorithm about your voice(labeled data) and it recognizes you afterward. The well-known applications are virtual assistants such as Siri, Alexa, Google assistant which works by hearing our voice commands.
3. Spam detection
This application is used where messages and e-mails are to be blocked. G-mail has an algorithm that detects strange keyword which could be fake such as “You win 5 lakh” or something and block those messages directly.
In this task, the data collected is unlabelled(undefined) and you can’t sure what your output will be. So, your algorithm understands the pattern from the unlabelled data and gives the required output.
As you can see in the above figure, in input there is a group of dogs and cats. At first, a child doesn’t see the dogs and cats, so he can’t recognize who is a dog and who is a cat? Here is no one to supervise him.
From time to time, he recognizes who looks similar to and different from another and categorize them.
After recognizes, he still don’t know the name of the animals, he just categorize them on the basis of their features.
This is how unsupervised learning works.
Got a question for us? Please mention it in the comment box.