I usually recommend to stay away from "data scienc...

# datascience

[bjonnh]

I usually recommend to stay away from "data science" books and instead focus on books on specific matters.

[dalexander]

Will do, thanks for the recommendation 🙂

[thomasnield]

@dalexander you should read my article to understand the rabbit hole of what you're asking : P https://towardsdatascience.com/how-it-feels-to-learn-data-science-in-2019-6ee688498029

[dalexander]

Oh, I’ve read your articles, I realized I was asking for something ambitious. I was hoping there was a “first steps into data science” book of some kind that was a survey of how to handle different kinds of problems that I could use as a reference so that when I encounter a new kind of problem I could go “Oh, this is an X problem, usually Y is used to address it” Like right now I’m reading about clustering data with high dimensionality (while waiting for the server team to fix an issue with the project I’m actually on), and I didn’t realize the level of complexity at first. So I started the process started by reading a paper about clustering and data structures for such, and thought a bit more about the problem, and then realized the paper was basically useless because it assumes only a few dimensions. Some of that could have been avoided if I had been exposed to a more broad survey of the field that I could reference when I run into a new problem. Strangely the “to data science, learn this” image in that article is somehow useful for that =p

[thomasnield]

@dalexander this is probably the best text for that purpose. https://www.amazon.com/Data-Science-Scratch-Principles-Python/dp/149190142X