| 21 Recipes for Mining Twitter |
|
Author: Matthew A. Russell Publisher: O'Reilly, 2011 Pages: 72 ISBN: 978-1449303161 Aimed at: Python programmers enthusiastic about the social web Rating: 4 Pros: Code-rich Cons: Comparatively expensive; you need to read the code Reviewed by: Mike James A spin-off from a bigger book, is this ultra-slim volume worth it? This is a very short and condensed introduction to using the Twitter API. You need to be very clear that it is just 60 pages of information and that might seem expensive given a cover price of $29.99, i.e. almost 50 cents a page! However, if expense isn't an issue, and it probably isn't if you have a really good idea that depends on mining Twitter, then this book might be the fastest way to get started. You have to try to keep in mind the fact that a big fat book is not necessarily the best value if it is stuffed and padded with irrelevance. You also need to know that this book is a spin off from Mining the Social Web by the same author - if you have the big book there is no reason to buy this one unless you value the code more than the explanation.
In this case the book hits the ground running and you are launched into Python examples from the first page. If you aren't up on Python then you choice is to learn the language or avoid this book. The format of the book is the usual cookbook format with each recipe setting out the solution to some, usually contrived, problem. This is a good format for this sort of topic and yes 21 recipes are about enough to see the API in action and to do just about everything you need. It starts off looking at OAuth and basically just connecting to Twitter and quickly moves on to actually retrieving data such as trending topics, searching for tweets, visualizing tweets and so on. Later it deals with the streaming APIs and how to dig into the data to analyze friendship relations say. Each of the recipes is fairly well explained, but there are longish listings without much description. In short, you need to read the code as well as the text to get much from the book. It also uses extras such as CouchDB and a range of Python modules. Overall, a good book as long as you are a reasonably competent programmer and are happy with Python. It isn't cheap but, if you have a focused aim to work with Twitter data, it will save you a lot of searching among the documentation. Related review:
<ASIN:1449388345> |
|||
| Last Updated ( Tuesday, 04 October 2011 ) |