An Introduction to Machine Learning

Hey,you ever wonder why the people are crazy about a booming term called "Machine Learning"these days? The following article will help us understand the Basics of Machine Learning.

In this article we will cover the following topics with some real-time examples.

1.What is MACHINE LEARNING?

2.Why is MACHINE LEARNING?

3.Types of MACHINE LEARNING

4.Advantages of MACHINE LEARNING

Let's Learn with fun.

What is MACHINE LEARNING?

Machine learning is an application of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. (Expert System)

That's the typical definition.Now,let us learn by real-time example.

Consider the following Scenario!

You are in a hurry rushing to your workplace.You want to book a cab through your smartphone.When the required destination is given in the app,it displays the fare amount.

Does the fare amount is always the same?Not in all cases...When you are using a platform like Uber,It shows the fares depending upon various factors such as peak time,bad weather,more demand etc.Now,does anyone are programming continuously sitting at the back end to provide you with new fares?Nope!!It's the concept of Machine Learning used there.

Let's see another example

Tony is a person who is a movie buff.He always like the action & comedy movies.He saw a movie named Mr.Bean which is a comedy film.Will he like that film?Yes.But if he see a movie called The Pain which is an action & romantic movie.Can you predict whether he likes it?With the given data it is not possible to predict accurately.Here comes the machine learning concept in predicting the new data based on the old data.

Data is "EVERYTHING"

Data is the heart of

Machine Learning

Without the data,we cannot accomplish machine learning!As humans,we do learn from our past experiences.In the similar way,machine learns from it's previous data.In nowadays,being everything recorded as data such as transactions,weather etc in the cloud/internet we can easily generate the humongous data.Although, systems now has more computational power.

Machine learning is a bit different from traditional computing.Here,the data & output is given to the system as inputs and we get program as output unlike traditional computing.

The program which is the output here,enables us to predict the future data based upon the data given as the input along with the expected output.

Why Machine Learning?

As the available data is getting huge day by day,implies it's possible to quickly and automatically produce models that can analyze bigger, more complex data and deliver faster, more accurate results even on a very large scale.

More or less,we are using machine learning everyday around us.It is obvious that with increase in number of users,data provided by them will make things easier in future.Literally,the increase in search efficiencies & automated recommendations in many of the websites is due to machine learning.

However,machine learning is the combination of statistics & computer science.

Hence,we can conclude that machine learning is used to analyse huge data in order to predict the future values which in turn makes life simple.

Types of Machine Learning

It can be broadly classified into 3 types.

1.Supervised Learning

2.Unsupervised Learning

3.Reinforcement Learning

Let us learn each by an example.

A,B,C are 3 friends who watches web series on Netflix.Consider 'A' likes web series such as Vampire dairies & Money Heist but dislikes the witcher.Similarly,'B' likes Money Heist but dislikes the witcher.Whereas,'C' likes the witcher & dislikes Vampire dairies.Now,A and B are kept near i.e.,considered alike in the database of netflix.But C is kept far away from both A,B since his interests are different compared to them.So,B would get recommendation of Vampire dairies to watch because the data near to him in the database is A who liked Money heist similar to B.This is "Supervised learning".Now guess,what recommendations would C get?

Supervised learning is the machine learning task of learning a function that maps an input to an output based on example input-output pairs. (Wikipedia)

Supervised Learning uses Labeled Data to train the model.Here,the machine knows the feature of the object & labels associated with those features.In machine learning, a target is called a label.A variable in statistics is called a feature in machine learning.

Unsupervised Learning

Suppose,we have a cricket dataset of various players containing number of runs scored & number of wickets taken.If we plot the data with runs on y-axis & wickets on x-axis,we can plot some data points.From that plot we can differentiate the data of players with more runs and players with more wickets forming two different clusters.So here we interpret these two clusters as batsman and bowler.

Unsupervised learning is a type of machine learning that looks for previously undetected patterns in a data set with no pre-existing labels and with a minimum of human supervision.(Wikipedia)

• The Learning of Unlabeled Data is Unsupervised learning.

Reinforcement Learning

Reinforcement Learning is defined as a Machine Learning method that is concerned with how software agents should take actions in an environment

• It works on the principle of feedback.

Advantages of Machine Learning

1. Easily Identifies Trends & Patterns which is used in Image Processing.
2. No Human Involvement is needed
3. Continuous Improvement
4. Handling Multi dimensional Data.

Some Real Life Applications Of Machine Learning

• Virtual Personal Assistants-say Google assistant,Siri etc.

Machine learning is an important part of these personal assistants as they collect and refine the information on the basis of your previous involvement with them. Later, this set of data is utilized to improve results that are used to your preferences.

• Email Spam and Malware Seperation

The system security programs that are powered by machine learning understand the coding pattern. Therefore, they detects new malware with 2–10% variation easily and offer protection against them.

• Search Engine Result Refining

Google and other search engines use machine learning to improve the search results for you. Every time you execute a search, the algorithms at the backend keep a watch at how you respond to the results.

That's we conclude our article on Introduction of Machine Learning.Hope You take some points home.Add any real time examples you encounter in daily life.