CRC Press Introduction to Data Science: Data Analysis and Prediction Algorithms with R

excellent

Very detailed, step-by-step guide to using R for data science. Certainly the most comprehensive text I've seen.

This book is a well-designed comprehensive introduction to data science using R. It guides the reader through a series of applied examples and explains how to think more effectively with R and especially the Tidyverse package to tackle real world data-related questions. Go for it.

One of the brief texts and concepts of data science with finest quality of papers. Also, the quick delivery is appreciated.

According to the author, those are notes to several Data Science courses from Harvard.The case studies are extremely interesting ranging from diseases, casinos, and even to Francis Galton’s height dataset (the birth of regression).My favorite was the wide use of Monte Carlo simulations.Don’t let the book title “Introduction to Data Science” fool you. While the author covers the basics, he dives into some advanced topics in full code (e.g. the Monty Hall or Birthday problem). Regarding “code”, he uses R code a lot and almost on any page (nice!).This “Harvard style” shows clearly, in my opinion. The author pushes the reader hard and, in my opinion, several topics (e.g. Moneyball) are not required for being a good Data Scientist. But, I believe, the goal is to push the reader to his limits.It’s a big book in hard-cover, but not in color.There are 38 chapters. With so many chapters, any reader should find very useful chapters.My personal highlights:- 9.5.2: Which base to use? Log10? Log2?- 13.2: Monte Carlo simulation for categorical data. In my opinion, the foundation for learning the exciting world of bootstrapping.- 14.2: How we can generate data on drawings from an urn. Here, you’re the owner of a casino and wonder, if you can make money with roulette.- 14.7: he uses a Monte Carlo simulation to show the Central Limit Theorem.- 16.5: Monte Carlo simulation for a disease.- 19.1 (spurious correlation): he runs a Monte Carlo simulation to create a synthetic dataset to show that we can find high correlations among uncorrelated variables.- 19.3 (reverse cause and effect): we can see that the David Robinson (creator of the broom R package) helped. Great linear regression (lm) done in tidyverse with the help of broom. An example I could not find in “R for Data Science.”Epic.FrancoPS: I love to buy great books and use highlights and notes. Yes, you can get the book for free, but I like to spend time off-screen. Yes, you can get the video lectures (same as text) for free, but videos are in general too slow for me. I believe most coders can read at 10X video playback speed.