SYSTEMS AND METHODS FOR RECOMMENDING TEMPORALLY RELEVANT NEWS CONTENT USING IMPLICIT FEEDBACK DATA

摘要

Computer systems, methods and computer readable media storing instructions, for providing news item recommendations are disclosed. An example system includes one or more digital memories having stored therein metadata for a plurality of news items and click data corresponding to user interactions with the plurality of news items, and a processor. The processor is configured to: determine a user profile for a user, the user profile including indications of news items previously clicked on by the user; select candidate news items for recommendation from (a) said news items based upon respective similarity distances to news items included in the user profile or (b) news items included in other user profiles that are identified based upon their respective similarity distances to the user profiles; and score the candidate news items using, at least in part, a temporal aspect of the candidate news items.