Recommending Discretion to Recommender Systems

You have seen recommender systems in action.  Visiting this blog will result in ads that are targeted to your preferences.  If you shop on-line, every item you view and/or purchase will trigger suggestions for other items that you may want to purchase.

These are powerful systems utilizing very complex and vast swaths of data to generate something that is personalized for you.

But have you ever wondered why some of the items that are recommended are, well, recommended?

One friend complained that upon creating a channel on Pandora, he was targeted with ads concerning:

  • Dating sites
  • Hair loss
  • Weight loss
  • Hardware stores

He did not require the first three items and can be safely described as not being a handyman.

Dad!  What's a testeroni sup mint?
Dad! What’s a testeroni sup mint?

So what selection of music would have concluded he was lonely, fat, bald and looking for home projects to occupy what must be the resulting quantity of free time due to his first three conditions?  And how does one not make a similar selection?!?

Needless to say, the ads missed their target.  And when playing the music out loud it opened some awkward questions from the kids who heard the advertisements.

The technology is very powerful but discretion algorithms could use a little work.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s