16
$\begingroup$

I am a self-taught web developer and am interested in teaching myself data science, but I'm unsure of how to begin. In particular, I'm wondering:

  1. What fields are there within data science? (e.g., Artificial Intelligence, machine learning, data analysis, etc.)
  2. Are there online classes people can recommend?
  3. Are there projects available out there that I can practice on (e.g., open datasets).
  4. Are there certifications I can apply for or complete?
$\endgroup$

closed as too broad by Sean Owen Dec 19 '15 at 9:25

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

15
$\begingroup$

Welcome to the site, Martin! That's a pretty broad question, so you're probably going to get a variety of answers. Here's my take.

  1. Data Science is an interdisciplinary field generally thought to combine classical statistics, machine learning, and computer science (again, this depends on who you ask, but other might include business intelligence here, and possible information visualization or knowledge discovery as well; for example, the wikipedia article on data science). A good data scientist is also skilled at picking up on the domain-specific characteristics of the domain in which they working, as well. For example, a data scientist working on analytics for hospital records is much more effective if they have a background in Biomedical Informatics.
  2. There are many options here, depending on the type of analytics you're interested in. Andrew Ng's coursera course is the first resource mentioned by most, and rightly so. If you're interested in machine learning, that's a great starting place. If you want an in-depth exploration of the mathematics involved, Tibshirani's The Elements of Statistical Learning is excellent, but fairly advanced text. There are many online courses available on coursera in addition to Ng's, but you should select them with a mind for the type of analytics you want to focus on, and/or the domain in which you plan on working.
  3. Kaggle. Start with kaggle, if you want to dive in on some real-world analytics problems. Depending on your level of expertise, it might be good to start of simpler, though. Project Euler is a great resource for one-off practice problems that I still use as warm-up work.
  4. Again, this probably depends on the domain you wish to work in. However, I know Coursera offers a data science certificate, if you complete a series of data science-related courses. This is probably a good place to start.

Good luck! If you have any other specific questions, feel free to ask me in the comments, and I'll do my best to help!

$\endgroup$
  • 1
    $\begingroup$ Coming back to this, Andrew Ng's course is hard. I should have mentioned I'm not strong in math. I've heard that this other Data Science course is a bit easier for learning the ropes. What do you think? $\endgroup$ – Martin Apr 1 '16 at 15:19
5
$\begingroup$

I am a self-taught data scientist, and I'd try my best to explain you how to go about it.


What fields are there within data science? (e.g., Artificial Intelligence, machine learning, data analysis, etc.)

Data Science is a very wide domain. It is about the science of data. So, any field which uses data to take decisions come under this domain. Some of the fields include:

  • AI
  • Pattern Recognition and Analytics
  • Bio-statistics
  • Statistical Learning
  • Machine Learning
  • Data Aesthetics(or data visialization)
  • Data Journalism

Are there online classes people can recommend?

I have answered a similar question. So I'd quote it here:

Start with the Coursera's Machine Learning course. It does a really good job in introducing the student to the domain of Machine Learning and helps you lay a solid foundation in the concepts.

In case, you feel that the math is a bit dumbed down in that course, you can take this course, taught by the same professor and is math-intensive than the former.

Now, you would have a clear intuition about the basic concepts of Machine Learning. Now, take this course, which can be said as a follow-up or a supplementary for the course of Andrew Ng.

This resource from IAPR has in-depth notes on a lot of ML concepts like cross-validation, regularization, etc.

You can also have a look at these amazing list of resources compiled into a blog on Quora.

Now, for diving into advanced concepts of neural networks and deep learning, you can make use of this free book.

Finally, the free e-book: Elements of Statistical Learning is a wonderful book for beginners in ML or Statistical Learning.

I addition to that, do check out this repository of data science references by Quora.


Are there projects available out there that I can practice on (e.g., open datasets).

I have started doing projects with open datasets of India. However, I would recommend you to check out this amazing discussion here, and after doing those projects, you can start with Kaggle.


Are there certifications I can apply for or complete?

In my opinion, there aren't any data science certifications. Yeah, there are a lot of Big Data certifications out there, but I didn't see them being really useful for a budding data scientist, so I recommend you not to chase them atleast till you are confident enough with your ML and data skills.

$\endgroup$
1
$\begingroup$

I recommend starting from Coursera specializations in data science. The data science specialization by Johns Hopkins is the oldest running specialization. I do not recommend books and kaggle. They only confuse you in the beginning. Keep in mind that coding is the easiest part of data science and you have to learn a lot. For getting an idea about the field, this Venn Diagram is a good start.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.