More Info

About the Similarities Engine and its author.

Overview

The Similarities Engine (SE) is a free music recommending system: pick five albums you love, and it finds listeners whose taste overlaps with yours and recommends the records they loved too. It uses no genres, tags, audio analysis, or AI model of what music is, only the accumulated pattern of which records keep showing up together in real people's lists of favorites. The method was first built as a DOS application in the early 1990s, served users by email through the late 1990s, and was granted US Patent 5,749,081 (1998, later cited by IBM, Google, Amazon, Microsoft, and Sony). What's live today is a faithful re-implementation of that original engine, running on modern serverless infrastructure and querying the original hand-grown data.

About the Author

David E. Whiteis is a computer systems engineer and serial inventor with a long history of throwing interesting technology pasta at the Internet wall and seeing what sticks. You can reach him by filling out the contact form here.

How It Works

Details on the technology behind SE, including the old and new versions, can be found here.