Method and apparatus for generating a recommendation for at least one content item

• US 10,237,604 B2
• Filed: 11/24/2006
• Issued: 03/19/2019
• Est. Priority Date: 11/30/2005
• Status: Expired due to Fees

• Current Assignee: Sisvel S.p.A. (Fineur International S.A.)
• Original Assignee: S.I.Sv.El. Societa' Italiana Per Lo Sviluppo Dell'Elettronica S.P.A. (Fineur International S.A.)
• Inventors: Pronk, Severius Petrus Paulus, Proidl, Adolf
• Primary Examiner(s): Lewis, Jonathan V
• Application Number:
  US12/095,001
• Publication Number:
  US 20080288982A1
• Time in Patent Office: 4,498 Days
• Field of Search: 725 46, 725 47
• CPC Class Codes: H04N 21/4147 PVR [Personal Video Recorder] H04N5/76 takes precedence ; arrangements for broadcast with accumulation-type receivers H04H20/40
• CPC Class Codes: H04N 21/4532 involving end-user characteristics, e.g. viewer profile, preferences monitoring of user activities for profile generation for accessing a video database G06F16/739; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18
• CPC Class Codes: H04N 21/458 Scheduling content for creating a personalised stream, e.g. by combining a locally stored advertisement with an incoming stream; Updating operations, e.g. for OS modules ; time-related management operations arrangements for replacing or switching information during the broadcast or during the distribution H04H20/10
• CPC Class Codes: H04N 21/4666 using neural networks, e.g. processing the feedback provided by the user
• CPC Class Codes: H04N 21/4668 for recommending content, e.g. movies
• CPC Class Codes: H04N 21/4756 for rating content, e.g. scoring a recommended movie

Abstract

A recommender engine (107_1, 107_2, 107_3) recommends further content items for playout on a personalized content channel on the basis of a profile (109_1, 109_2, 109_3) associated with that personalized content channel. The profile (109_1, 109_2, 109_3) includes an indication of the user'"'"'s “like” and “dislike” of content items played out on said personalized channel.

Claims

1. A recommendation apparatus for generating a recommendation for at least one content item, comprising:

a filtering unit associated to a plurality of personalized content channels and configured to filter content items complying with predetermined criteria associated with the filtering unit, wherein only content items selected by said filtering unit are played out on said personalized content channels, wherein the filtering unit includes a plurality of filters and each filter is associated with a corresponding one of the plurality of personalized channels;

a recommendation engine unit connected to an output of said filtering unit and associated to said plurality of personalized content channels, wherein said recommendation engine and said filtering unit operate in series, wherein said recommendation engine unit is configured to calculate a recommendation score for said content items previously selected by said filtering unit, wherein said recommendation engine unit includes a plurality of recommendation engines, wherein each recommendation engine is associated with a corresponding one of the plurality of personalized channels such that each filter is associated with a corresponding recommendation engine and personalized content channel;

a profile store device for storing user profiles, wherein each of said user profiles is associated to a corresponding one of said plurality of personalized content channels;

a user interface;

a processor operating and coupled with said profile store device, said filtering unit, and said recommendation engine, the processor executing the steps of, without requiring a user to log in;

filtering each content item using said filtering unit such that only content items that meet predetermined selection criteria of the filters are selected for playing out on said corresponding personalized content channels;

forwarding a list of said selected content items, for each of said filters, to said recommendation engine unit;

calculating a recommendation score using said recommendation engine unit for said selected content items on the basis of said user profile associated to said corresponding personalized channel;

a scheduler configured to;

receive a list of content items selected by said filtering unit and the corresponding recommendation scores from said recommendation engine;

schedule each content item selected by said filtering unit on the basis of the received recommendation scores; and

select at least one content item for subsequent playout according to the schedule;

wherein the processor is further configured to;

determine at least one of a like and a dislike of a content item being played on said corresponding personalized channel by monitoring actions taken by a viewer through said user interface regarding said content item as said content item is being played on the personalized channel; and

update the user profile stored in said profile store associated with said personalized channel on the basis of the determined at least one like and dislike associated with said content item being played on a corresponding one of said plurality of personalized channels, said update including updating each attribute of the content item with one of the at least one like and dislike such that an understanding of a user'"'"'s preferences in the plurality of user profiles increases over time.

2. The recommendation apparatus according to claim 1, wherein the step of filtering each content on a corresponding one of said personalized channels such that only contents that meet the predetermined selection criteria are plated out on said personalized channel comprises filtering each content with a Boolean filter.

3. A recommendation apparatus for generating a recommendation for at least one content item, comprising:

a filter associated to a personalized content channel and configured to filter content items complying with predetermined criteria associated with the filtering, wherein only content items selected by said filter are played out on said personalized content channel;

a recommendation engine connected to an output of said filter and associated to said personalized content channel, wherein said recommendation engine is configured to calculate a recommendation score for said content items previously selected by said filter, wherein said recommendation engine operates in series with said filter;

a profile store device for storing a user profile;

a user interface;

a processor operating and coupled with said profile store device, said filter, and said recommendation engine, the processor executing the steps of, without requiring a user to log in;

filtering each content item using said filter such that only content items that meet the predetermined selection criteria are selected for playing out on said corresponding personalized channel;

forwarding a list of selected content items to said recommendation engine;

calculating the recommendation score for the selected content items in the list based on the user profile associated to said personalized content channel;

a scheduler configured to;

receive the list of content items selected by said filter and the corresponding recommendation scores from said recommendation engine;

schedule each content item selected by said filter on the basis of the received recommendation scores; and

select at least one content item for subsequent playout according to the schedule;

wherein the processor is further configured to;

determine at least one of a like and a dislike of a content item being played on said personalized content channel by monitoring actions taken by a viewer through said user interface regarding said content item as said content item is being played on the personalized content channel; and

update the user profile stored in said profile store associated with said personalized content channel on the basis of the determined at least one like and dislike associated with said content item being played on said personalized content channel, said update including updating each attribute of the content item in the user profile with one of the at least one like and dislike.

4. The recommendation apparatus according to claim 3, further comprising:

a filtering unit, the filtering unit including a plurality of filters;

a recommendation engine unit, the recommendation engine unit including a plurality of recommendation engines;

wherein the profile store unit stores a plurality of user profiles,

wherein each of the plurality of filters is associated a corresponding one of the plurality of recommendation engines and a corresponding one of the plurality of user profiles.

5. A recommendation apparatus for generating a recommendation for at least one content item, the apparatus comprising:

an information store device configured to store electronic program guide data that is retrieved from at least one source, the electronic program guide data associated with content items;

a filter associated to a personalized content channel and configured to filter the content items by applying predetermined criteria to the electronic program guide data, wherein only content items selected by said filter are played out on said personalized content channel;

a profile store device for storing a user profile;

a recommendation engine connected in series to an output of said filter, connected to the profile store device, and associated to said personalized content channel;

a user interface;

a processor operating and coupled with said profile store unit, said filter, and said recommendation engine, the processor executing;

filtering each content item using said filter to generate a list of selected content items such that only content items that meet the predetermined selection criteria are selected for playing out on said corresponding personalized channel;

forwarding the list of selected content items to said recommendation engine;

calculating a recommendation score for the selected content items in the list based on the user profile associated to said personalized content channel, wherein the filter, the user profile and the recommendation engine allow personalized content to be recommended for playing on the personalized content channel including when the user profile does not have a sufficient understanding of a user'"'"'s preferences;

a scheduler configured to;

receive the list of content items selected by said filter and the corresponding recommendation scores from said recommendation engine;

schedule each content item selected by said filter on the basis of the received recommendation scores; and

select at least one content item for subsequent playout according to the schedule;

wherein the processor is further configured to;

determine at least one of a like and a dislike of a content item being played on said personalized content channel by monitoring actions taken by the user through said user interface regarding said content item as said content item is being played on the personalized content channel; and

update the user profile stored in said profile store associated with said personalized content channel on the basis of the determined at least one like and dislike associated with said content item being played on said personalized content channel, said update including updating each attribute of the content item in the user profile with one of the at least one like and dislike, wherein the user profile changes over time such that the recommendation score calculated by the processor change over time as the understanding of the user'"'"'s preferences increases.

6. The apparatus of claim 5, wherein the recommendation is generated without requiring a user to log in.