Machine Learning is Fun – Part 5 – Probability Distribution Concept

Share
Viewer Rating

Hi Friends! I hope you liked my previous posts. Today, I am going to talk on an important topic; that is Probability Distribution. Well, it’s like you are trying to assume or predict something in a scientific way.

Now the question is Probability on what? Variables obviously!

First concentrate on 2 kind of variables : 1. Discrete (Example like coin toss, rolling die, counting person etc.)

2. Continuous (Example like height measurement, price distribution etc.)

Now the question is why do we need to learn probability distribution?

Because to measure something which is very important in our daily life, i.e.  RISK Assessment which is applicable in every aspect of our life.

To understand it scientifically, one must have good knowledge in probability distribution.

Next, if I give you some sample of T-shirts and ask you to bring near to similar kind of size shirt from market, what shall you do?

Obviously you shall take average .. in statistics, we use Mean.

Next, if I give you multiple color balls and ask you “what is my favorite color “.. you shall pick the most common color among the balls.

In statistics, it is called Mode.

If I ask you to predict the best price of flat in an area ..Where flat prices like 1k, 1.5k,2k,2k,3k,2.5k,100k.

What would be your answer?

2k.

This is called Median ( I hope you shall not consider the sudden higher flat price which is 100k and very odd).

The above terms are called central of tendency of a distribution.

Now the next thing is if I have some samples and I draw a graph of those samples’ distribution, what would be the other parameter I need along with central of tendency?

The answer is variability; that means how the samples are distributed or spread from the center. This is called Variance or Standard Deviation.

I hope you like my post. Stay tuned for more!!

2 thoughts on “Machine Learning is Fun – Part 5 – Probability Distribution Concept”

Leave a Comment

%d bloggers like this: