Good Learning Resources for Data Scientists

I  get asked a lot of times what are the learning resources that I go to,  to learn more about the Data Science or Artificial Intelligence. This  blog post will allow me to share with others easily on the resources I  depended upon for my learning and how I used these learning resources. I  will start off with the websites that I go to.

Arxiv.Org

Journals  used to be a very closed domain and every academics fight tooth and  nail in order to get their research published in the most prestigious  (or known as Tier 1 in the academic circle). The existence of these  journals make working in the university a lot more attractive to me  then, as compared to now. But the audience that these journals reached  out to is very limited and given the current circumstances where  research is not limited universities anymore, research articles need to  be shared more widely in order for humankind to advance.

This is where Arxiv.Org comes in. Its housed under Cornell University Library and they are  treasure troves of research done by a lot academics and non-academics.  You’ll find a lot of research over there and you can practically learn a  lot about data science, algorithms, computer science etc from Arxiv.  The key to using it is to be able to search with relevant keywords.  Otherwise an alternative is perhaps FB groups or Reddit, where people  “recommends” certain research papers and you can read based on  recommendations.

I  do admit that the papers can be challenging to read but you have to  tell yourself that these papers are not meant to finish within one  reading. You have to read it a few times to appreciate and make  connections with what you have learnt in previous readings and other  relevant knowledge gained previously.

So  brace yourself if you want to be in Data Science and Artificial  Intelligence. Arxiv.Org will be a great companion in your learning  journey.

StackOverflow

You cannot run away from programming given such a large movement in open- source and especially large tech firms open up their machine learning frameworks, like Tensorflow, PyTorch, Keras etc.

Since you cannot run away from programming, you cannot run away from the most  common and frustrating task of a programmer and that is debugging! For anyone who codes, StackOverflow is the place to go to and look for fellow comrades who has fallen to the same bug before and seek solutions. I do go to StackOverflow often (Yes, I am a lousy programmer.) to search for possible ways to debug and most often I do get some ideas on how to debug from there. The big challenge is trying to decipher based on the “titles” of the question, whether the solutions you are looking for is right over there. So most  of the time, you have to look through several Q&A threads before you  find the possible solution.

Blogs

Blogs have came a long way since it started (about 10 years ago?). Blogs is not simple as just a place for people to write about opinions but they are now “products” such as publications where it consolidates blogs of similar topic so that new bloggers like me have a ready audience easily.  The ready audience encourages more people to contribute their opinions  and ideas.

Medium used to be a great place to learn until it starts charging for content. I used to read a lot of Medium articles but these days I am very turned off by them as all they recommend are articles that requires Medium subscription. There are good authors in them but apologies, there are just too many free and good contents out there to be digested so Medium posts are brought to the end of the queue. There are good blogs out there, so in order to keep abreast of the development, may have to tap onto RSS Feed and an RSS reader to do it.

MOOC

I picked up a lot of knowledge from MOOC sites, such as Coursera, (especially) Udacity and Edx. I loved these sites given the tremendous amount of knowledge and materials provided by universities around the world.

Imagine in the old days, such advanced modules are not available online and if I were to learn about it, I have to asked around for a recommended textbook and then eat bread (cheap ones) for about a month, before I saved enough money for the textbook. Now, thanks to MOOCs  sites, I can have access to these curated materials easily.

One thing to note is that most of the modules has videos that are very concise so do not expect to be an expert after going through the  modules. What it can give you is a list of keywords that you will need  to research further in order to master the topic.

YouTube

Another life savior and a big thanks to the universities and tech meetup/ conference for putting up the lectures and talks on YouTube! I  loved it! YouTube these days has a lot of universities lectures  available and if I need a quick explanation on certain topics, I will go to YouTube to look for a short video on it and played it at 1.5–2X  (another fantastic feature!).

World-renowned  universities such as MIT and Stanford put their lectures online and allows anyone who has Internet access to “sit in” them and learn together! How fantastic is that? So I keep a list of playlists that I hope to go through and learn more about computer science/data  science/artificial intelligence. If video is your thing, there is another website you can go to. Its called VideoLectures. Here is the link. It contains videos of lectures and presentations at academic conferences so there can be nice surprises in them.

Podcasts

During time of commute, I usually follow up on podcast to find the latest news of Data Science and Artificial Intelligence. In fact, there are times where they come up with good explanation on some of the difficult concept in Machine Learning or Artificial Intelligence, useful explanation that I can use for my training. Podcast also allow me to listen to different opinions from experts and gives me the possibility to learn during my long  journey to meetings. Its great relaxation to listen to the knowledge  sharing of other experts in the relevant fields with the scenery passing  by. I’ve picked up a lot of knowledge during these bus journeys and  they provide a good exposure to new concepts as well. I believed that these exposures are due to the spontaneity of the panelist/speakers. I am in the midst of preparing my own podcast show so do drop by from time to time.

If you are looking for a list of good podcast in Data Science and Artificial Intelligence, I think this post (from KDD Nugget) might help

Conclusion

So these are the sites that I go to to learn and pick up more knowledge on  data science and artificial intelligence. No doubt these are sites you probably have heard of, but my sharing here is more of creating an awareness that you can learn more about data science and artificial intelligence from there.

I hope the blog has been useful to you. If it has, do share it. Have fun in your data science learning journey!