I am very excited about using this module. One question - can this module be used with User Relations, as an example "your friends/coworkers/etc - browsed/flagged/rated this node/term/entity"?
Thank you for the module it brings a whole bunch of new options.

Comments

danithaca’s picture

danithaca commented

You can do this without using Recommender API though: just simply aggregate the browsing history of your friends and then you can get the recommendation for you. Although RecAPI could definitely facilitate the "aggregate" process. If more people are interested, I can write a helper module.

Recommender API is designed for more sophisticated cases where it first guesses which users have similar tastes as you based on browsing history etc, and then make recommendations based on similar tastes.

mefisto75’s picture

mefisto75 commented

Thank you for the reply. I don't really know how to aggregate friends' browsing history) sorry
As far as I understand there are several ways of "recommending" content or rather try to get useful info out of the flow. That is by using existent modules like Similar by Terms, using Views to create related pages listings, using Flag/Fivestar to filter out what was marked by other people as worthy. I have my hopes high for RecAPI. Wating for proposed cloud service to try it. Maybe wrong.
I am to build a site where people are offered content (from news thru RSS feeds to book reviews) based on their tastes (preferences?)
It's obviously becoming more and more important to filter information out somehow while waiting for semantic web.
Like what these guys did http://tweetedtimes.com/

danithaca’s picture

danithaca commented
Issue summary: View changes

This might require a customized helper module to take user relation data and make recommendations. If you are using a Drupal module that handles "user relation" data, please post a request to implement a RecAPI sub module at #2377503: Upgrade Recommender API helper modules to be compatible to 7.x-6.x.

danithaca’s picture

danithaca commented
Status: Active » Closed (works as designed)