Jeffrey M. Stanton

This is a very nice, gentle, not too technical introduction to data science for non-specialist.

Beware though: if you are an instructor and want to use this as a text for your class, review all the R code very carefully. I did, and I had to do extensive work as some package functions were deprecated and no longer function (chapter 10), in the chapter on word clouds, the function designed by the author to extract the text from tweets did not work properly (a lot of misclassified tweets). In the chapter on databases, no indication was made anywhere that in order to use one's machine as local host, you need extra software such as Wamp. In the chapter on association rules, one package was not mentioned but its absence prevented another package from functioning properly, at the same time that package was not properly installed as dependency and had to be installed manually.

All of this not necessarily a big deal but do not just send your students to this text without warning or additional instructions.

My solution: create functioning R scripts for all the chapters that needed them, with R code corrections and extra instructions where needed.

Important winning point of this text: it's free. Not a trivial consideration as far as I'm concerned.

… (lisätietoja)