Source: Deep Learning on Medium
In this article we are going to learn about six elements of machine learning before you start coding the machine.Everyone just tries to jump into doing stuff without knowing what they want and what they have in hand.Here I am going to simplify things from what i have learnt.
Okay let’s start..Before jumping into machine learning,Have you heard of all Jargon surrounding the machine learning?Let’ look at a Jargon cloud.
Do you know about all these Jargons? and when and how to use it? then you are an Expert.For those who are new to these things,i will walk you through with a simple example(i may be wrong).
I hope everyone knows about cricket,Let us assume you are the captain of the indian cricket team and you need to select the 11 players(Jargon) out of 15(Jargon cloud) for the match against austraila(Machine Learning).
What does the captain(you) do?
Before selecting 11 players he should know the strength and weakness of all the 15 players,when to use them and when not to use and who is the best to achieve his objective.
Just think for a moment that all the Jargon you see in a Jargon cloud are players and you(captain) need to decide what jargon you need to use and how do you organise it properly so that you know which part of machine learning you are dealing with and that is of interest to you or you rather focus on other element of machine learning.
Now we will look at how all these Jargon fit into these Jars.Let us try to fill all the jars with each of the six elements of machine learning.
All the jars are empty as of now,later in this article we should be able to get some sense out of the each jar.
What do you think the fuel for ML?
Data: yes,you guessed it right,the data is the fuel for machine learning.we all know that data is everywhere and we experienced what companies can do with our data right? Have you heard of google and facebook?ohhh nooo,common man….now you have an idea right what am i talking about?
Every company has lot of data.Lets have a look at amazon website for a one plus 6 mobile.
You can see that there is so much of data which is spread in different category like specifications,reviews by the customers and in many ways. We have facebook at the other end which has lot of data, which takes customer information and they will tell you where and when you went to the particular place and time.I know it’s scary,but how do they do it.You will be thinking that oh god they have so much of data,let me tell you..
Have you experienced when you searched for some product in some website and switch the website to amazon,you would see that your searched product is shown by amazon under recommended products.How do they do it. That’s where machine learning comes into the picture.You feed all the data to the machine learning so that it learns and data and find a relation between the given input and output.But the data to be in the structured format.Let’s have a look at the one of the format.
First column shows finger print scan of different persons and x is the structured format of the scan and y is the output whether it has certain characteristics or not.
We have successfully filled the first jar with huge Data in it.In the next article we will fill other jars too.Stay tuned…