blogs | wanderingstan

I saw through my MyBlogLog Widget that Stan had stopped by my site.   He’s the guy that thought up Lijit.  It’s cool that he came to my blog because I mentioned Lijit in a post just a couple hours old.  And then I find HIS blog with the quote below. “When Amazon tells you that 63% of people who bought product X also bought product Y, there is nothing social involved. This is math.

I couldn’t agree more.  Pure numbers does not equal social – it just equals numbers.

A nice article over at Read/Write Web about recommendation engines. They break down current systems into 4 categories: 1. Personalized recommendation – recommend things based on the individual’s past behavior 2. Social recommendation – recommend things based on the past behavior of similar users 3. Item recommendation – recommend things based on the thing itself 4. A combination of the three approaches above I take issue with the label “Social recommendation” because that’s not what social is. Social means being part of a society, being connected. When Amazon tells you that 63% of people who bought product X also bought product Y, there is nothing social involved. This is math.

Source: blogs | wanderingstan


  1. I’m not saying that by supplying that number, Amazon became a social networking site. Only that that number is, in itself, and by identifying common interest, an aspect of the much broader social networking paradigm. I also don’t think that social networking (or, broader, social behavior) is restricted to only verbal communication. Dressing in “goth” outfits, having a “mohawk” hairdo, “not-farting-in-public” etc etc are all expressions of social behavior as well. So is watching the same movies and reading the same books. While calculating a percentage of people who have common interests in itself doesn’t constitute social behavior, it certainly addresses an aspect of said behavior.

  2. You can mean different things by the word “social”. And sure, what Amazon does could be compared to what sociologists do with studying behavior patterns. But in the Web world, “social” means social networks — explicit connections between people who interact with each other. And it’s these explicit connections that Amazon knows nothing about, but which mean so much to me as an individual. Now that these social connections are available online, they should be used for recommendations. *That* will be the next breakthrough in recommendation systems.

    [And hey Rob, thanks for using Lijit. Please ping me with any ideas/comments/frustrations!]

  3. I don’t quite agree (Rob pays me for that!).
    That’s saying that “90% of all marine species became extinct during the Permian extinction” has nothing to do with natural history, it’s just math. Or that “52% of the American peple voted democratic” has nothing to do with politics, it’s just math.
    There IS a social aspect to that ‘Amazon’ number.
    It’s about common interests. Isn’t that what social networking tries to do? Bringing people together based on shared interests? Sure, Amazon’s GOAL is commercial in nature, but it uses (cleverly) aspects of social networking.
    “Social” doesn’t necessarily have to involve actual communication .. a sense of “belonging to” and “commonality” is as much an aspect of it too.
    I, for one, ALWAYS look at what people who shared my interest in a certain book, are also interested in. That and the reviews have lead to many additional book purchases.
    I think it’s silly to start yelling “it’s just math” when you spot a number somewhere…as soon as there is a context.. it’s no longer “just math”.