Data Mining the Amazon, 2003

Book using’s recommendation algorithm to determine if polarity in partisan politics is present in music tastes.

Slideshow of the book pages

Download pdf version >>

“The use of bar codes in commercial outlets, the computerization of business transactions and the explosion of e-businesses provide a flood of data to consumers and corporations. For years companies have been analyzing this data and extracting information to boost sales. This process, known as data mining, gives this book its name. The Amazon referred to is the online warehouse of books CDs movies, toys, etc. which has developed some of the most sophisticated filters for their vast database. Based on these filters, Amazon makes recommendations to their customers.” –excerpt from book

CD recommendations for books read by self-identified conservatives

At the time of this research (1999-2001), had public user-generated reading lists called Listmania!. The books that seeded the music tastes charts in this book were from these Listmania! reading lists created by self-identified political conservatives and liberals.

When people still purchased CDs on Amazon, any book might have a list of CDs that were also purchased with the book. “People who bought this book also bought these CDs” information for each of the Listmania! books was translated into large network diagrams for each ideology in an attempt to draw out distinctions in taste.

Inspired by annual report aesthetics, the book features multiple charts and graphics that not only outlay music tastes across the political spectrum, but also niche reading lists and purchase behaviors for various corporations and organizations.

CD recommendations for books read by self-identified liberal customers

52 pages, 2 fold-out charts
Dimensions (height x width x depth): 23 x 14.5 cm.
paperback, offset printed, perfect bound
Edition: 500

Design: Angie Waller and Alex Ching

2 Responses

Comments are closed.

Back to Top