bemore

What does bemorehuman do?

bemorehuman is a recommendation engine; sometimes these things are called recommender systems. It recommends things such as books, or movies, or beer, or music bands to people. You input implicit or explicit ratings data or purchase data or click data, and bemorehuman will build a model of that dataset. With the model built and loaded into RAM, you can then query the system via a REST interface and receive recommendations in real-time.

Why is it useful?

We don't want to waste time getting poor quality recommedations. Often, when the word "recommendation" is used what's really meant is "popular thing for your demographic." If you're lucky, that is. If you're not lucky, recommendation can easily mean "different models of the thing you just bought" or "whatever we're paid to promote this week." Our software takes a different view. We think it's most important to give recommendations purely based on what the individual receiving the recs might like. This way, the traditional marketer-focused mechanisms can be bypassed and focus can be given to what people would find interesting or useful.

How it works

Here is a simple example of how bemorehuman works:

• What do you want to recommend? Books? Movies? Classified ads? Things that work well are things that are unique, don't change, and have a persistent id. And of course you need the legal right to use that data for your specific purposes. For example, "red wine" is a good category of things to recommend because there are lots of different red wines such as "Seifried Nelson Pinot Noir 2020" which is a unique product or thing. In the installation instructions below we show an example of how you can recommend movies.

• The input to bemorehuman is a list of 3-value anonymous events. An event would be like: "person 2637 gave product 312 a rating of 4." So the input file would look like:

  2637  312  4  <-- this is the event mentioned above
  311   3363 2  <-- next event
  1288  778  5  <-- next event
  ...and so on
  

• The ratings (it doesn't have to be ratings; we can use purchases, clicks, etc.) file from the previous bullet is used as input to the valence generator, valgen. A valence is just a pairwise relationship between two things. The valence indicates strength or weakness of connectivity between those two things. We use simple statistical methods like linear regression and correlation coefficients to determine the bond between two things.