Back to results
US 2013/0097153 A1Published application
Profile Based Content Retrieval for Recommender Systems
AssigneeAxel Springer Digital Tv Guide GmbH
InventorsMauro Barbieri, Serverius Petrus Paulus Pronk, Jan Korst
Life of this patent
Estimating the term…
Abstract
The present invention relates to an apparatus, a method and a computer program product for controlling a recommender system, wherein a user profile normally used by a recommender to predict user ratings is employed to generate a targeted query for the remote database yielding a set of results that can be scored by the recommender and provided as suggestions to the user.
Claims7
An apparatus for controlling a recommender system, said apparatus comprising:
- a) a feature-value extractor (116) for analyzing a user profile (109) and for extracting feature-value pairs which characterize a user of said recommender system;
- b) a query composer (118) for composing based on feature-value pairs extracted by said feature-value extractor (116) a query to be sent to at least one content source (103);
- and
- c) a recommender engine (107) for rating results obtained in response to said query so as to provide a recommendation for said user,
- said apparatus being characterized in that it further comprises a feature selector (117) for selecting from said feature-value pairs extracted by said feature-value extractor (116) those feature-value pairs which are most discriminative for said user profile (109), wherein the feature selector (117) is adapted to use the Relief algorithm as a feature selection algorithm for selecting said most discriminative feature-value pairs.
2 more elements
A method of controlling a recommender system, said method comprising:
- a) analyzing a user profile and extracting feature-value pairs which characterize a user of said recommender system;
- b) composing based on extracted feature-value pairs a query to be sent to at least one content source (103);
- c) rating results obtained in response to said query so as to provide a recommendation for said user;
- d) iterating said composing and rating steps until a predetermined number of relevant content items has been obtained;
- and
- e) counting co-occurrences of feature-value pairs that appear in a same item and combining only those feature-value pairs with a sufficient co-occurrence count in said query.
3 more elements
Dependencies are inferred from the claim text.