What books for a beginner on Data Science?

Question

Books

Decided to learn ukzanoe in tags. I have not yet decided what exactly, but as I understand it, we must begin with the theory of probability and / or mathematical statistics. I studied both subjects at the university, but I have already forgotten everything. Tell me what book to read. Now there is a choice between K. Dougherty "Introduction to Econometrics" and Gmurman "Probability Theory and Mathematical Statistics" - can someone tell the fundamental difference between these books?

Answer 1

The question is somewhat strange in view of the really INCREDIBLE number of books on DS,
BD, ML released recently. In any language, for any starting level of education, with an eye on various tools. Take any and start learning. When something becomes clear and you want to deepen it, or vice versa, when something becomes incomprehensible, you purposefully look for another source in which the topic already known to you is set out deeper or clearer. And so you move, expanding your horizons in the topic.
If it is difficult to find it yourself, there are already compiled lists, for example:
https://ru.stackoverflow.com/questions/Books-and-learn...
Or
https://www.learndatasci.com/free-data-science-books/
And there are thousands of tips on the net - how, what and in what order to learn. For example:
https://proglib.io/p/learn-data/
Here are a bunch of additional resources.
www.7wdata.be
https://www.datasciencecentral.com
https://datascienceplus.com
https://www.kdnuggets.com
https://www.analyticsvidhya.com
https://towardsdatascience.com
Good luck.
PS "There is a choice between K.Dougherty's "Introduction to Econometrics" and Gmurman's "Probability Theory and Mathematical Statistics" - can someone tell me the fundamental difference between these books?"
----> Fundamental difference: Gmurman gives a general theory, a good foundation, a universal textbook.
Dougherty is more focused on a specialized niche of economic and social problems.
Both are not bad for studying statistics in their respective specialties at the university.
I would not recommend starting an independent study with them. Take a look at the list and recommendations above and choose from them.

Answer 2

on the root tracker for the request "big data" (sorted by time added)

Answer 3

If everything is fine with the understanding of Ang, then I advise you to start with the course from Stanford . If you want to learn, you will understand what knowledge in mathematics you lack, and you will be able to learn specific topics.

Answer 4

https://medium.com/machine-learning-for-humans/how... - from simple to complex. Need English.

Answer 5

You need to know Python well, first of all, and solve problems, and not learn theory for DS. Libraries: Numpy, Pandas, Matplotblib, Scikilearn, Seaborn etc. especially.
If you really want hardcore mathematics, then immediately read the book from the creators of this science ISLR by Gareth James (available for free on the university website in pdf).
I really enjoyed Jose Portilla's lectures on udemy.com. It seems that there are still lectures by citizen Eremenko that are not bad on the topic, but I only took a course on Tableau from him.
"Learn Data Analysis with Python" by Henley, Wolf: started now, but these are only puzzles and what employers require for employment in essence, not spatial theory.