Intelligent system and methods of recommending media content items based on user preferences
First Claim
Patent Images
1. A method of predicting items of media content most likely to appeal to a user, comprising:
- accepting ratings, at a multimedia recording device, of media content items from a user, via a graphical user interface, according to a rating system;
expressing the rating system using a graphical metaphor;
receiving by the multimedia recording device, from a server, correlation data between pairs of available media content;
generating ratings at the multimedia recording device of media content items not rated by the user according to one or more predictive algorithms, based on the media content items specifically rated by the user and the correlation data received from the server;
displaying, by the multimedia recording device, the media content items ratings by the user and the generated media content ratings in the graphical user interface; and
allowing the user to correct generated media content ratings displayed in the graphical user interface in order to obtain generated media content ratings more in line with what the user expects.
12 Assignments
0 Petitions
Accused Products
Abstract
A system and method for making program recommendations to users of a network-based video recording system utilizes expressed preferences as inputs to collaborative filtering and Bayesian predictive algorithms to rate television programs using a graphical rating system. The predictive algorithms are adaptive, improving in accuracy as more programs are rated.
-
Citations
51 Claims
-
1. A method of predicting items of media content most likely to appeal to a user, comprising:
-
accepting ratings, at a multimedia recording device, of media content items from a user, via a graphical user interface, according to a rating system; expressing the rating system using a graphical metaphor; receiving by the multimedia recording device, from a server, correlation data between pairs of available media content; generating ratings at the multimedia recording device of media content items not rated by the user according to one or more predictive algorithms, based on the media content items specifically rated by the user and the correlation data received from the server; displaying, by the multimedia recording device, the media content items ratings by the user and the generated media content ratings in the graphical user interface; and allowing the user to correct generated media content ratings displayed in the graphical user interface in order to obtain generated media content ratings more in line with what the user expects. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15)
-
-
16. A method of predicting items of media content likely to appeal to a user, based on the user'"'"'s preferences, comprising:
-
rating a new item based on earlier ratings of items by the user that correlate highly with the new item according to a first predictive algorithm when the new item has earlier ratings of items by the user that correlate highly with the new item, the first predictive algorithm comprising a collaborative filtering algorithm; rating the new item, when the new item has not been rated by the first predictive algorithm, according to a second predictive algorithm, the second algorithm comprising a content-based adaptive modeling algorithm, based on ratings of specific features by the user describing the item, wherein at least a portion of the features have previously been explicitly rated by the user; for a new item whose features have not been previously rated by the user, rating the new item based on previous explicit ratings of items by the user, according to the adaptive modeling algorithm, wherein the adaptive modeling algorithm generates implicit ratings of features based on the explicit ratings of items by the user; and allowing the user to correct a rating for an item generated by the first predictive algorithm or the second predictive algorithm in order to obtain generated item ratings more in line with what the user expects; wherein the method is performed by one or more computing devices. - View Dependent Claims (17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37)
-
-
38. A method of predicting items of media content likely to appeal to a user, comprising:
-
predicting a rating of an item based on features explicitly rated by a user; in the absence of explicitly rated features, predicting a rating of an item based on implicitly rated features, wherein an adaptive modeling algorithm generates the implicit feature ratings based on previous overall ratings of items made by the user; and allowing the user to correct a predicted rating for an item in order to obtain predicted item ratings more in line with what the user expects; wherein the method is performed by one or more computing devices. - View Dependent Claims (39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51)
-
Specification