Data science is one of the more vague terms thrown around in the business world. Does it mean analytics, statistical modeling, machine learning, database architecture, or software engineering? Depending on the company and job description, it could mean or all of the above. Do you need a degree or even a PhD? Yes… And no.
One of the limitations I had with starting data science was myself. I felt like because I didn’t have the credentials, I could not (or would not) be allowed to do the work. However, since I started learning the craft I have found myself (and others) who knew just as much as people who came out of formal programs. And this is not to be blasé towards people who have Masters or PhDs – there are many, many areas where their expertise has shown through and they have rightfully earned their place in the upper echelon of Data Science. What I mean is just because they have an advanced degree, it does not mean you cannot participate or contribute. And if you have an advanced degree, I am sure I don’t have to tell you that you will still need to keep learning as well as rely on others.
In all of this, data is a language, and the way you understand, manipulate, and communicate data is a constant journey. There will always be new things to learn, new areas to explore, and new technology to keep you on your toes. To take a recent example, Quantum Computing is gaining traction, which will change how Data Scientists tackle really difficult, computationally expensive problems. Quantum Computing is based in quantum physics, which is truly a new language for most of us. (I mean, I still don’t feel like I’m at an advanced level of linear algebra and calculus, and now you’re going to throw quantum entanglement at me?!).
With all that being said, there are a million resources for getting started. There are formal education programs (Masters and PhDs), e-courses, books, and even Quora posts about everything you can study. Everyone has their flavor of “how-to” and which things to do or avoid. But it’s a journey. I have some of my favorite things listed below, but note that these are resources and not guides. I am more than happy to tell you what I’ve done, but I think if you set yourself up on these different platforms, you will quickly find what is recommended and which place to start. The most important thing is to just start.
Social Media:
- Reddit Subs – excellent community of people helping each other
- Twitter accounts – many times I just bookmark these or read when I have time
- LinkedIn Accounts – Great for throwing in some learning on your newsfeed.
Studying:
- OReilly Books ($)
- Kindle Books (textbooks; $)
- Medium ($)
- Arxiv (Research papers)
- Researcher App
Coursework:
- Pluralsight ($)
- DataCamp ($)
- Coursera ($, but you can audit!; Also, Andrew Ng’s Machine Learning course is frequently mentioned as the best course for ML.)
- LinkedIn Learning ($)
- Kahn Academy
- Kaggle (Not just competitions – they have lessons as well!)
- YouTube (depending on who/what you follow, of course; Also, very much worth paying for premium to avoid ads)
- Krish Naik
- Edureka!
- StatsQuest
- Crash Course Computer Science – For those of us without a CS degree, this gives you a TON of great information on how computers actually work and how they were developed.
Coding Examples:
Others Helpful resources:
- Note Taking
- YouTube – productivity
- Crash Course Study Skills
- Thomas Frank’s channel
- Evernote
- YouTube – productivity
- Productivity
- Habitica – Create healthy habits to study
- Toggl time-tracking – keep track of your time and “bill” yourself for your work
- Anki – Create flash cards from what you have learned and keep studying (even on the go!)
- Pomodoro timer (or just use your phone’s timer)
- Extra Curriculars – Reinforce your learning!
- Start a blog
- Start a YouTube Channel
- Journal your learnings each day
($) – subscription-based, but many libraries and employers may offer this for free!
That’s a ton of things to check out, but out of all of them, my main go-to’s are Medium, Coursera, O’Reilly, YouTube, and Evernote:
Medium is amazing for keeping up on current trends in the industry as well as learning fundamentals. Everyone and their grandma (including yours truly) has posted stuff to Medium. I frequently bookmark articles to read later or even use as tutorials for new methods and concepts.
Coursera is Coursera – it’s such a widely known e-learning resource that it needs no introduction. However, my little “hack” for Coursera is to audit whatever you can, whenever you can. It’s great to get the certificate, but if you just want knowledge, then you can skip the quizzes and apply what you are learning. Projects > Certificates, in my humble opinion.
O’Reilly’s library of books have been monumental in helping me understand different concepts. In fact, most of the heavier Data Science books I would buy are offered with the subscription to O’Reilly.
YouTube is life. Want a quick 5 minute tutorial on NLP? It exists. Need a 40 hour course on Deep Learning? That exists, too. And it’s not just no-names doing this. Many of these channels are ran by universities or professors.
Evernote is my “aggregator” of knowledge. You can see how I use this as part of my Data Science Cookbook. I scribble notes, type up code concepts (one of the few notebook apps I have that uses code blocks, which feels absurd these days), and even write up my blog posts in Evernote. Much like you wouldn’t go to school without your textbooks, you certainly wouldn’t show up without a notepad and pen.
My list is not perfect by any means. I am curious to hear what you think. What are some of your go-to resources? What has been a game changer in your mind?
Fuzzy Logic Blog 