In this tutorial, we will be focusing on what is machine learning, what are its applications, and the classification.But before learning machine learning, you should have a basic knowledge on

  1. Probability and linear algebra
  2. Programming skill – especially in Python.
  3. Graph theory
  4. Calculus – especially derivatives of single variable and multivariate functions

What is Machine learning?

a small Learning bot

As the word machine learning suggests it enables computers to learn from experiences and past data on their own. And the definition given by Arthur Samuel in 1959 is –

“Machine learning enables a machine to automatically learn from data, improve performance from experiences, and predict things without being explicitly programmed.”

A Machine Learning system use algorithms that learns from historical data, builds the prediction models, and whenever it receives new data, predicts the output for it. As we give more information, it gives higher performance.

If we are distinguishing between traditional programming and machine learning, we can say that in traditional programming we give input and program so that it gives the output. Whereas in machine learning we give the input and output so that it learns from the source and give us the program.

difference between traditional programming and machine learning

Applications of Machine Learning

Imagine you want to purchase a headphone and you searched on amazon. Next time when you open the website you will find many headphones recommendation. Have you ever wondered why? It’s because of machine learning that is, it analyses the user’s interest and recommend the product.

Likewise, there are many applications for machine learning such as biometrics, video recommendation, spam filtering, language translation and much more.

Machine Learning - Robots learns from Humans

The need for machine learning is increasing day by day because it can help with complex issues. We humans have limitations like we can’t access huge amount of data, we can’t do a work repeatedly etc. So we give huge amount of data so that it learns, constructs and predict the result.

Classification of machine learning

Machine learning can be classified into 3 –

  • Supervised learning
  • Unsupervised learning
  • Reinforcement learning
classification of machine learning

Let’s look forward on each classification –

Supervised learning

In this type, we focus on mapping the inputs with the outputs. 

The machine is given labelled input data so that it creates models based on labelled data, learns each data and predicts the output.

Supervised learning is further divided into classification and regression.

types of supervised learning


If you want the machine to recognize between a male and female employee. We give some features as input data for each male and female. The objective of the classifier will be to assign a probability of being a male or a female (the label) based on the information (features given). Later when the machine learns, we can test by giving new data. So that it predicts the result.


Prediction of rain using temperature and other factors is an example. Here the machine develops a relational model between dependent and independent variables using independent variables.

Unsupervised learning

In this type, we focus to restructure the input data into new features or a group of objects with similar patterns.

Here the input data (without any labels or categorization) is given and the machine learns without any supervision.

Unsupervised learning is further divided into clustering and association.

types of unsupervised learning


A technique in which the data is divided into groups such that objects with the possible similarities remain in a group that has less or no similarities with another group.

Grouping of documents according to the topic is an example of clustering.


In this technique, it checks for the dependency of one data item on another data item and finds some interesting relations among the variables of dataset.

An example is Market Based Analysis.

Reinforcement learning

In reinforcement learning, the agent interacts with the environment, explores it and as a result it gets a reward for each right action and gets a penalty for each wrong action. The goal of an agent is to get the most reward points, and hence, it improves its performance.

Keep looking forward for more tutorials…!!

PyCharm installation guideline for Windows OS

Starting Python Project in PyCharm IDE

Debugging Python Project in PyCharm IDE :