Trust me you don’t need to believe me. The best way is always the one you find and not the one you copy.It had been a year and I had been just dancing on this concern. I wanted to learn but neither had money to spend nor any proper direction. Everyone told me one thing or the other. Well, First of all, I would like to classify the whole Data Science in four parts; Data Acquisition, Data Preprocessing, Data Analysis and finally Prediction.
Machine learning or Deep learning comes in the last part and to get to it you must have some fundamental ideas of how prior steps are taken. The first is about mining data from various sources using various techniques, the next is about trimming the excess and filling the vacancies with some expected or certain values. The second last is analyzing it and abstracting meaning out of the crude data. And the last part is about using the data for predicting outcomes with some different possibilities.
The way I used to learn them is somewhat different. I learned first to use Excel sheets, make calculations and statistics using the Excel sheets. This helped me understand what is actually happening with the data inside of it. Do you know the best part about Excel? You could manage the first three parts with just excel, you don’t have to even learn to code for that. After all data science has been already in use but not as we call it to know.
The next is to learn to code. There are many programming languages for it. I personally prefer Matlab. It comes with a price but with a million of features along with it. It is a programming tool for Engineers and Scientists.Using Matlab is not like learning to code. Most of the calculation and data manipulation is done with simple maths and logic operation. All you need to know is Maths and statistics, the rest is covered and you can rest assured of it. To learn the basics there are ONRamp self-paced Courses available on the Mathworks site. You can learn a lot from there.
Well, this was enough to know what is Data Science, but if you wish to make a career into it, you need to learn to code and get a bit professional. Polish your skills with Python and you are gonna make a lot of cash out of it. There is a not so popular course available on Coursera, you should try that first before going with popular courses like those on Udemy, Udacity or Andrew Ng. The name of that course is Machine Learning by IBM. They teach it with love and I’m sure you are gonna enjoy it.