Comments of dplass (2)

Comment on post progress on the netflix front:

I think I understand. But what i was thinking about more was to correlate the users and not the movies. In other words, "find users who are like me" and ask "how did *they* rate this movie?" It's collaborative filtering (see where the steps are:

1. Look for users who share the same rating patterns with the active user (the user who the prediction is for).
2. Use the ratings from those like-minded users found in step 1 to calculate a prediction for the active user

Comment on post progress on the netflix front:

(Full disclosure: I'm "working" on the Netflix prize too).

All I've done is some thinking (and loading the data into mysql)

What do you mean by "calculating the correlation between every pair of movies using the dot product"? Specifically, "correlation" and "Once I have that data it should be fairly straightforward to apply that to all the other movies a user has rated and come up with a new rating" - how?

If you don't want to reveal your thoughts/algorithms, that's cool.

This backup was done by LJBackup.