Machine learning is the branch of AI. Machine learning is based on the concept that machines and systems can analyze and understand data, and learn from it. Machine learning makes decisions with minimal to zero human intervention. Most industries and companies that process large amounts of data have realized the value of machine learning technology. Machine learning process the data science Course and businesses are able to work more efficiently and gain an advantage over others. Above all machine learning is a self-learner itself. In 2020 there are more corporates and SME’s who have felt the importance of machine and machine-based programming so there is more scope for professionals in the IT industry who has more and continues development in the knowledge of Artificial Intelligence.

Types of Machine Learning


Supervised learning is often described as task-oriented. Supervised machine learning is focused on a single task with multiple examples to the algorithm until it gives an accurate performance on that task. Applications which are examples for this supervised machine learning are :

  • Advertisement Popularity: Supervised learning task in selecting advertisements is that it will perform well. As you browse the internet you come across many advertisements which have been recommended by the learning algorithm that they were of reasonable popularity. As a result, the expense and time for the trial and error are saved.

  • Spam Classification: The spam filter in the email system is a supervised learning system. Such emails which are filtered as spam or not spam are learned by the system as malicious emails or not so that the user is not harassed by them. It also saves on chance of hacking confidential data and in turn leads to the security of system and data.

  • Face Recognition: Do you use Facebook? Face recognition is used by Facebook which is an example of supervised learning algorithm that is been trained to recognize your face. Having a system that takes a photo, finds faces and guesses who that is in the photo (suggesting a tag) is a supervised process. It is safe for users to use social media as a platform to market products, services with the knowledge of the account owner.

Looking forward to becoming a Data Scientist? Check out the Data Science Bootcamp Program and get certified today.


Unsupervised learning is very much the opposite of supervised learning. Unsupervised learning works on the data and its properties; we can say that unsupervised learning is data-driven. The outcomes from an unsupervised learning task are controlled by the data and the way it’s formatted. Some common examples are:

  • Recommendation system: If you have ever used YouTube or Netflix, you will most likely encounter a video recommendation system. These systems are usually placed in unsupervised domains. We understand the different categories of video duration, type, language, etc. We also understand the viewing history of many users. Taking into account users that have watched similar videos like you and then enjoyed other videos that you have yet to see, a recommender system can see this relationship and prompt as per the data as a suggestion. Hence you have a choice of movies to select as per your like. It saves time and you have a mood buster as you don’t have a search on choices. 

  • Purchase habits: Purchase habits are recorded in the database and used in unsupervised learning algorithms to group customers into similar purchase segments. This helps the company to target these grouped markets and can even resemble a recommendation system. Companies can also stock products based on buyers’ habits and fast-growing products. The buyer can also choose according to his preferences. When he visits the portal, this option will appear in the buyer’s top list.

  • Group user logs: There are fewer users, but they are still very relevant. We can use unsupervised learning to group user logs and questions. This will serve as a reference for the company to correct problems that customers face in order to improve products or design answers to frequently asked questions. Either way, it is a proactive task. If you have ever submitted a product issue or submitted a bug report, you are likely to use it in an unsupervised learning algorithm to rank other similar problems. Therefore, it can improve the delivery system of products or services.

  • Reinforcement: Reinforcement learning is fairly different when compared to supervised and unsupervised learning. Where we can easily see the relationship between supervised and unsupervised (the presence or absence of labels), the relationship to reinforcement learning is a bit murkier.

Reinforcement learning as per my knowledge is learning from mistakes. If I place a reinforcement learning algorithm into any environment it will make a lot of mistakes initially. When we provide some sort of signal to the algorithm that associates good activity with positive signal and bad activity with a negative signal. We can reinforce our algorithm to prefer good activity over bad ones. And this would over the time make our learning algorithm make fewer mistakes as it made before. Hence the system itself learns and improves the quality of service.

Where is reinforcement learning in the real world?

Video Games: One of the most common places to look at reinforcement learning, it is in learning to play games. Look at Google’s reinforcement learning application Alpha Zero and Alpha Go which learned to play the game Go. Our Mario example is also a common example.

Reinforcement learning agent deployed as a game AI is a good option for game developers to take help.

  • Industrial Simulation: For many robotic applications, it is useful to have machines learn to complete their tasks without having hard-code their processes. This option is safer and cheaper as we can incentivize our machine to use less electricity that saves money.  Hence the research and development cost as well as time to use permutation and combination will be saved.

  • Resource management: Reinforcement learning is ideal for navigating in complex environments. We can meet the needs of balancing certain requirements through resource management. In Google’s data centre, they use reinforcement learning to balance the needs of our electricity needs and proceed as efficiently as possible, thereby cutting the main cost. How does this affect us and ordinary people? The cost of cheap data storage has less impact on the environment shared by all of us.

Machine learning in 2020 can be Supervised or Unsupervised. If you have a lesser amount of data and clearly labelled data for training, opt for Supervised Learning. For large data sets, we generally use unsupervised learning to give better performance and results.

Also Check this Video

Recommend Reads:-

Recommended videos for you

Join the Discussion

Interested in Henry Harvin Blog?
Get Course Membership Worth Rs 6000/-
For Free

Our Career Advisor will give you a call shortly

Someone from India

Just purchased a course

1 minutes ago

India Address:

Henry Harvin House, B-12, Sector 6, Noida, Uttar Pradesh 201301

FREE 15min Course Guidance Session: