How To Identify The Best Resources To Be A Self-Taught Data Scientist

By Kimberly Cook |Email | Jul 24, 2018 | 14223 Views

Let me start by saying that there is no one way to being a self taught data scientist. You have got to find out what works for you.

On that note here's what I would like to provide you in this guide.

  • How to identify what you want to be (and finding the right resources)
  • Resources I use and what works for me

Identify The Skills You Need
So you know you want to get into Data Science but don't know where to start. Here are some pointers to get started.

  1. Take a look at a couple of Data Science related job descriptions and see what appeals to you. Would you like to be a Machine Learning Engineer, a Data Engineer, a Data Analyst or perhaps a Data Business Analyst?
  2. Next make a list of all the skills you find in the job descriptions. See if you already have any of those skills. Perhaps you already know a little SQL or Python that can be leveraged or maybe you are good at Data Analysis.
  3. Next find the gaps in your skills and identify which are the most popular skills among all the job descriptions you have been looking at. Pick about 5 new skills and start working from there.

Identify Resources
Now that you have the new skills you have to learn, you need to identify resources that work for you.

  1. Google: Here use all your googling skills..I know you have them.. :-) . Something like 'Top resources to learn xyz' should do the trick.
  2. Quora: Another great option to look up for advise.
  3. Class Central: Here you can look up any MOOCs that you have identified and check their ratings/feedback from students who have taken the class. This will help you avoid wasting time on classes that might not suit your learning style.

Just Dive In! I cannot emphasize this enough. Once you know the basics the best way to learn is to get started and code.

  1. Kaggle: It's my go to source for datasets. There are a vast variety of datasets from beginner to advanced level. You can also take part in competitions to see how you rank among others.
  2. UCI Machine Learning Repository: Another great source for datasets that are segregated by data type, machine learning model etc.
Resources I use and what works for me
Here's a list of courses I love and keep going back to for reference time and again.


Other things that work for me

  • Kaggle datasets: Find a dataset you like and start working on it. At first it will only be summarizing and cleaning the data but slowly you'll understand how to apply machine learning models.
  • Twitter: There is a huge inspiring community on twitter where you can find the latest happenings in the Data Science world. Also there are a couple of learning challenges that occur weekly/monthly, explore and find one what you like. Some of my favorite areâ??-â??#TidyTuesday #SoDS18 #MakeoverMonday and #100daysofMLcode

While these resources may or may not be the right choice for you, I hope it gives you a starting point on how identify what Data Science field works best for you and how to get your unique way of learning started.

The article was originally published here

Source: HOB