Collaborative filtering

I wanted to learn a little more about how sites such as Amazon compile their suggested product lists based on other people’s spending habits.

I came across this lecture delivered  via a MOOC by the University of Washington. The lecture clearly explains that it (broken down into basic form) is simply a case of counting the number of shoppers that bought any combination of products and offering the most popular items as last minute add-ons.

I found this very helpful when considering the building blocks of algorithms. It would appear that in this instance “Big Data” is being used for creative analysis for the benefit of the masses (Enyon 2013). However there is certainly scope for acknowledging that there are other behaviours within an online shopping experience that may not be identified by spotting trends.

I hope to cover such behaviours and trends by conducting a small experiment and documenting my findings before the end of this week.


Enyon, R. (2013). The rise of Big Data: what does it mean for education, technology, and media research? Learning, Media and Technology 38(3): pp. 237-240.

Weapons of Math Destruction: how big data and algorithms affect our lives by Guardian Science Weekly

“Algorithms are trying to service us based on a computer profile of what we do online, that only has to do with what we do online. It has nothing to do with what we do offline”

This quote has been ringing in my ears since listening to this podcast.

I define an algorithm as a computed process that generates outputs based on trends, statistics and behaviours. Trends and statistics can often be clean cut. However I wonder how accurate digital representations of human behaviour can ever really be?

The consequences of getting it wrong will only continue to grow as algorithms are increasingly embedded into our daily lives.
via IFTTT & Soundcloud