How to start learning machine learning / deep learning / data mining?

Question

data mining

Good afternoon!
Where to start learning machine learning , given that experience in this area is 0?
If this area has many directions, I would like to delve into AI.
Are there any blogs and books (preferably in Russian)?
It is also desirable that the examples in the articles / book be in one of these languages ​​(in descending order):
* JS
* Python
* Java
* Ruby
* C
R, Prolog, etc. - they are not interested, since these are narrowly focused languages.

Answer 1

T. Segaran - programming the collective mind. There are some machine learning methods out there. The book itself is a mixture of the web and Data Mining. Python
Theory in these two books:
Peter Norvig. Artificial intelligence: a modern approach
Book P.S. Romanov. Fundamentals of Artificial Intelligence
The scikit-learn library with implemented algorithms from the books above. Python
An article on it on Habré: Introduction to Machine Learning with Python and Sc...
And it will also come in handy (at later stages)
Kozlov M.V., Prokhorov A.V. Introduction to mathematical statistics. - M .: Publishing house of Moscow State University, 1987.
Borovkov A.A. Math statistics. Estimation of parameters, testing of hypotheses. — M.: Nauka, 1984
Borovkov A.A. Math statistics. Additional chapters. - M .: Nauka, 1984.
In the early stages, it is desirable to know the course of theory and mathematical statistics from school. What is probability, variance of magnitude, etc. - these concepts are sometimes used. Especially widely in genetic algorithms
PS I myself am still learning, the list is from a person who has been engaged in AI for a long time

Answer 2

List of books, videos and courses on machine learning and mathematics, all in Russian. Large, high quality selection. Almost all pdf'ki of books are googled.
https://ru.stackoverflow.com/a/683632/1084
do not listen to Arseniy Kravchenko , it is quite possible to learn well in Russian, but yes, you will need to follow the latest trends.

Answer 3

https://habrahabr.ru/company/mlclass/blog/252743/ is quite understandable and, in my opinion, reasonably written.
I went through many of the courses listed before this article was written. But he also made a lot of extra moves.

Answer 4

https://github.com/ujjwalkarn/Machine-Learning-Tut...
But in general English and Python/R are required. You won't get far in Russian and Javascript.

Answer 5

Sebastian Raschka - Python and Machine Learning
Brink H., Richards D., Feverolf M. - Machine Learning

Answer 6

In addition to the courses already mentioned (from Andrew Ng and Yandex / MIPT on Coursera), I can recommend two courses that I took myself:
* Introduction to Computer Science and programming from ... - learn to write code in Python, learn about some types of algorithms and their properties. You solve problems. It helped me learn how to write code and understand what information to look for in order to study further. The course is free.
* Data Science program at Yandex.Practice- A solid introduction to DS and machine learning. Of the benefits - good applied business cases, calls with a mentor (practicing DS) and projects that are checked by a person. Of the minuses - some topics are given superficially (for example, working with big data and Spark). The course is paid, with a free introductory part - you can try it - it's yours or not.

Answer 7

I will also advise deep learning in python by Francois Chollet and support the answer about Yandex workshop https://praktikum.yandex.ru/ . In addition, there are a bunch of other courses and tutorials. mlcourse from ods is praised.