0
$\begingroup$

Nowaday, there are lots of online courses related to machine learning, and there are lots of programming languages which can be used to do data mining, such as R, Python or Matlab. For example there are lots of data science related courses on coursera, some use R and some use Matlab.

Lots of the time I just download the videos to my disk, and never open it again, other times I spend all my time to finish the courses and just forget the real problem I want to study. When I am using machine learning tools, I find it's better to know some underlying mathematics, then I would pick up some book such as convex optimiztion, but this time I find it's too slow to master such materials. Seems there is no plan at all when I study machine learning techniques.

So anybody here who can share something about your methodology when you try to do machine learning or data mining, do you have systemic and effective plans?

$\endgroup$
1
$\begingroup$

The approach that I took (mathematician (PhD) by trade with interest in statistics and nowadays as everyone: machine learning):

Go to https://www.kaggle.com/ and pick one of the self study projects (e.g. the Titanic challenge). There are tutorials where you learn essential techniques applied to the problem at hand.

Additionally you can read the following two books that are free to download: The Elements of Statistical Learning: An Introduction to Statistical Learning

Recently I read this survey article Big Data: New Tricks for Econometrics and it does both: the Titanic data set and it references the free books.

$\endgroup$
1
$\begingroup$

Try to get a copy of "Machine Learning: A Probabilistic Perspective" by Kevin Murphy. It is well written, having good coverage with depth. It will give you a nice feel for the topics.

I also like David Sontag's approach. All the reading materials are based on freely available materials: http://cs.nyu.edu/~dsontag/courses/ml14/

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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