The easiest way to do this is substitute the user-id with the item-id and vice versa. Do not change the names used in the input JSON or in the queries.
Where the input is documented as (user-id, event-type, item-id) just swap the ids, the input may ask for user: user-id but send item-id instead. Where the query is for user=user-id just send an item-id. This will work for only a single event—the conversion event/primary event. You train a model to find items preferred by users, and with an item-id in the query you can get “recommended” users. They will be called items in the json, but ignore that, the JSON is just a label. On Nov 7, 2016, at 12:52 PM, Magnus Kragelund <[email protected]> wrote: Hi, I'm using Prediction IO with the Universal Recommender engine template (https://github.com/PredictionIO/template-scala-parallel-universal-recommendation <https://github.com/PredictionIO/template-scala-parallel-universal-recommendation>) to generate personalized and item based item suggesttions, on a web site selling event tickets. I feed user interactions like pageviews, ticket purchases, rating etc to Prediction IO . The setup is great for displaying event suggestions to the site visitors, but I'd also like to be able to have Prediction IO suggest a list of users, that would be most interested in a given item. With that in hand i would know who to contact, to sell out events that are not already sold out. How would I go about getting such suggestions from Prediction IO, using the Universal Recommender - or any other - engine template? Best regards Magnus Kragelund
