So this is brain dump. So it might get boring or feel like rant. Anyways let’s start
What exactly is data science?
It is a detective work as far as I know. You get the data and then the detective work starts, you clean the data so that there is some consistency in the data and after that you visualize it, then comes your observation skills and then comes ‘trial and error’ of using various models and figuring out which might be the best for you.
-Sounds easy but it is not.
The work takes a lot of time during each phases, learning something new every time and somehow there is always some problem in the initial phase, may it be the break down of some package similar to how you break down at night, or it might just be compatibility issue with some code your team mate sent akin to the relationship you are pursuing.
You do a lot but hardly anything ever gets noticed because you are judged based on the solution you provided. Ahem, does that ring any bell.
SO, Why do it anyway?
If it is done right you will get to learn a lot a about a domain or business you are working on, the amount of real world knowledge and inferences you can get out of the data science and the real world solutions you can make will give you that high which you might be searching for in your drugs and alcohol. It is not immediate but again if you work for others and feed them the solution you cannot hope to feel the effect of drugs for yourself. you need to work on ‘your’ problems. The problems that you feel are worthy or interesting.
For working on your problems you will need a dataset though and data science is boring because everyone is busy working on a dataset provided to you but no one is willing to make a dataset for them self which they will enjoy working on.
For example, I will have so much fun with data regarding the footfall and kind of footfall my locality has with time stamp and other parameters. This can help in my decision to open a café near my place a very practical endeavor.
Now you will say Pankaj, why don’t you do it yourself, it will be so much fun.
No, I cannot do that because I am lazy, data collection is a hassle and I would rather have someone just give me that data on a silver spoon. Argh, Bring it to me already.
And this is where I should end this because it is becoming more of a rant than a brain dump.
Thank you for reading this.