AI Products 

What is machine learning?

Tushar Mahajan
What is machine learning?

This guide is for anyone interested in machine learning but doesn't know where to begin. I'm sure many people tried reading the Wikipedia article, became frustrated, and gave up, wishing someone would just give them a high-level explanation. That is exactly what this is.

Because the idea is to be accessible to everyone, there are a lot of generalizations. Who cares, though? If someone becomes more interested in ML as a result of this, mission accomplished.

What is machine learning?

Machine learning is based on the premise that there are generic algorithms that can tell you something interesting about a piece of data without requiring you to build specialized code tailored to the situation. You give data to the generic algorithm instead of writing code, and it creates its logic depending on the data.

A classification algorithm is an example of an algorithm. It can categorize data into many groupings. Without modifying a line of code, the same classification system that recognizes handwritten numerals could be used to classify emails into spam and not spam. It's the same algorithm, but it's fed different training data, therefore the categorization logic is different. To know more about Machine Learning visit us: machine learning course in Pune.

"Machine learning" is a broad word that encompasses a wide range of generic algorithms.

Two kinds of Machine Learning Algorithms

Machine learning algorithms can be divided into two types: supervised learning and unsupervised learning. The distinction is subtle but critical.

Supervised Learning

Assume you work as a real estate agent. Because your company is expanding, you hire a group of fresh trainee agents to assist you. But there's a catch: you can look at a house and have a pretty decent notion of what it's worth, but your trainees don't have your experience, so they have no idea how to price their homes.

You decide to build a small software that can assess the value of a house in your area based on its size, neighborhood, and what similar houses have sold for to aid your trainees (and possibly free yourself up for a trip).

So for the next three months, you keep track of every time someone sells a house in your city. You write down a lot of information for each house, such as the number of bedrooms, the square footage, the neighborhood, and so on. Most importantly, you make a note of the final sale price:

We want to construct a program that can estimate the value of any other house in your region using that training data:

It's referred to as supervised learning. You knew how much each house went for, so you already knew the solution and could work backward from there to find out the logic. To know more about Machine Learning visit us: machine learning classes in Pune.

You send your training data on each house into your machine learning algorithm to create your app. The algorithm is attempting to determine what type of math is required to make the numbers work.

This is akin to erasing all the arithmetic symbols from a math test's answer key:

Can you figure out what kind of arithmetic problems was on the test based on this? You understand that you must "do something" with the numbers on the left to obtain each response on the right.

You let the machine figure out that relationship for you via supervised learning. And after you've figured out what math was needed to solve this particular set of issues, you'll be able to answer any other problem of the same type!

To know more about Machine Learning visit us: machine learning training in Pune.

Tushar Mahajan
Zupyak is the world’s largest content marketing community, with over 400 000 members and 3 million articles. Explore and get your content discovered.
Read more