Machine Learning - know it right! Know it all... Part 1

Setting up the stage!

Replay - Let’s go 15 years back, when mobile phone was infrequent. You might have seen; only classy people were using Internet. World is changing very fast. Technology change is very rapid in this time. Now mobile phones and Internet are available in every pocket.

We are probably living in the most defining period of human history. The period when computing moved from large mainframes to PCs to cloud. But what makes it defining is not what has happened, but what is coming our way in future.

Inquisitive about! what is machine learning?

Hold on, before taking a deep dive into machine learning – lets feel something, lets feel the presence of machine learning around us.

Let me give you a common example which is obvious to most of us – but we hadn’t thought about it, Really!

Wonder why you see relevant add all over?

Ok! Let me explain…

Let’s do some online shopping today, probably Amazon would be a good choice (lol, Amazon --- I’m advertising for you, here).

Any ways you get on the search and identify a product of your interest, suppose you are looking to purchase a mobile phone, you added that to your cart or Wishlist.

At this very point you changed your mind and your out of Amazon website -

After that, started searching something on internet.

Hmmm! That blog sound interesting and you go-ahead and open it, oh my God the ad is coming on this website is the same as your Wishlist product.

How this magic is possible! 

Come on we live in the world, where magic is just a myth… Agree, Hmmm! Yes, you do.

Yes! Your mind is rightly shouting out ----- 'Machine learning Algorithm'

Background! Please.

Machine learning is a subset of AI. That is, all machine learning counts as AI, but not all AI counts as machine learning.

For example,

·        Symbolic logic

·        Rules engines,

·        Expert systems

and

·        Knowledge graphs

could all be described as AI, and none of them are machine learning. Clear…

One aspect that separates machine learning from the knowledge graphs and expert systems is its ability to modify itself when exposed to more data; i.e. machine learning is dynamic and does not require human intervention to make certain changes. That makes it less brittle, and less reliant on human experts.

Let me bring you back to the point highlighted above.

“With machine learning, product features are only possible if the data permit them.”

Note: Data is key – Completely different topic and its colossal subject, Some other time, please.

Coming back.

Now that we have fair understanding of Machine learning – lets put on you Diving Gear.

Yes! Your right we are going deep dive.

Before let me give you an example on why ML?

Handwritten digits are a classic case that is often used when discussing why we use machine learning, and we will make no exception.

Below you can see examples of handwritten images from the very commonly used MNIST dataset.

The correct label (what digit the writer was supposed to write) is shown above each image. Note that some of the "correct” class labels are questionable: see for example the second image from left: is that really a 7, or actually a 4?

MNIST – What's that?

Every machine learning student knows about the MNIST dataset. Fewer know what the acronym stands for. In fact, we had to look it up to be able to tell you that the M stands for Modified, and NIST stands for National Institute of Standards and Technology. Now you probably know something that an average machine learning expert doesn’t!

In the most common machine learning problems, exactly one class value is correct at a time. This is also true in the MNIST case, although as we said, the correct answer may often be hard to tell. In this kind of problems, it is not possible that an instance belongs to multiple classes (or none at all) at the same time. What we would like to achieve is an AI method that can be given an image like the ones above, and automatically spit out the correct label (a number between 0 and 9).

The roots of machine learning are in statistics, which can also be thought of as the art of extracting knowledge from data. 

That’s a good one, Art – let me present to you an art that was shared by a very beloved one…

The Buzz is real! Man.

Types of Machine Learning:

“Wait, so you mean there’s more than one way in which a machine learns?”

A little intimidating, but don’t worry, this next bit doesn’t get harder (it’s really easy to understand, I promise!).

Type 1 Supervised Learning:

The first type of machine learning is supervised learning. And it’s exactly what the name sounds like (see! it’s already so simple). This type of learning is when the developer labels the variables that the machine will be working with. 

How it works: The first thing to keep in mind, in order to avoid big mistakes. Split your data set into two parts: the training data and the test data.

This algorithm consists of a target / outcome variable (or dependent variable) which is to be predicted from a given set of predictors (independent variables). Using these set of variables, we generate a function that map inputs to desired outputs. The training process continues until the model achieves a desired level of accuracy on the training data.

In this style of Machine learning, we give some training to machine based on that we test the accuracy of Machine Learning Algorithm on testing data. Suppose We have made Price prediction algorithms based on machine learning and we have ten years past data of that Stocks. Now we can use three years data for training to machine and rest 7 years data can be used to test and analyses the accuracy of our algorithm.

Hope your with me and everything said makes sense! 
                                          Nooo? you might be kidding – Hmmm!

Whatsoever, within this domain there are two sectors of learning: Regression and Classification.

Regression is the machine’s ability to recognize numbers, and group them together to form predictions. An example of these variables can be the total area of a house in square feet, the number of bathrooms it has, and the number of bedrooms it contains. Through linear regression, the machine is able to predict the cost of a house by grouping different examples of houses and learning from their variables and costs.

Ok, that makes sense.

Classification is the machine’s ability to identify images, or things that are binary (yes’ and no’s). Think of this like playing flashcards with your machine! Your stack of cards contains different types of cats and dogs. The side you (the developer) can see is the image, and the machine sees the back of the card. Using numbers in between 0 and 1, the machine tries to guess what the animal is. It keeps doing this until it is able to distinguish between cats and dogs.

Now you understand – The First part of ML, have energy to jump on to next.

See you in part 2 of Machine Learning.....