Back to results
US 8,286,206 B1Granted
Automatic rating optimization
AssigneeAT&T Intellectual Property I, L.P.
InventorsJeffrey Aaron, Thomas Anschutz
Life of this patent
Estimating the term…
Abstract
Automatic rating optimization is described. In an embodiment, ratings of a program can be received from one or more rating sources. Based on these ratings, a representation of a content selection mechanism can be sent to potential consumers of the content. Access events for the content can be counted over a duration of time so a determination can be made regarding how the ratings provided by each of the rating sources affect popularity of the content. A weight accorded to ratings received from each of the rating sources can be adjusted based on the determination. Profiles can be established for consumers and/or rating sources.
Claims15
A method for optimizing ratings associated with content using a tangible, non-transitory content rating computer system, the method comprising:
- the content rating system receiving ratings, the ratings being associated with the content and being received from rating sources;
- the content rating ascertaining whether the rating sources accessed the content for which the ratings were received;
- the content rating system determining an effect each of the ratings has on popularity of the content, the determining comprising:
- analyzing, for each rating, data indicating a number of times the content is accessed after each of the ratings is received by the content rating system;
- evaluating a speed of change in the popularity of the content over a duration of time, the speed of popularity change being represented as:
- S=(Pop2−Pop1)/(Time2−Time1),
- where S is the speed of change, Pop2 is a popularity of the content at Time2, Pop1 is a popularity of the content at Time1;
- and
- evaluating an acceleration of the change in the popularity of the content over the duration of time, the acceleration of popularity change being represented as:
- A=(S2−S1)/(Time2−Time1),
- where A is the acceleration of popularity change, S2 is a speed of popularity change at Time2, and S1 is a speed of popularity change at Time1;
- the content rating system weighting at least one of ratings, the weighting being based, at least partially, upon results of the determining and ascertaining steps, wherein:
- the weight of the rating is increased if the rating comprises a positive rating and the popularity of the content increases;
- the weight of the rating is increased if the rating comprises a negative rating and the popularity of the content decreases;
- and
- the weight of the rating is increased if the rating source accessed the content;
- the weight of the rating is decreased if the rating is a negative rating and the popularity of the content does not decrease;
- the weight of the rating is decreased if the rating comprises a positive rating and the popularity of the content does not increase;
- and
- the weight of the rating is decreased if the rating source did not access the content;
- the content rating system generating an average rating for the content, the average rating comprising at least two weighted ratings;
- and
- the content rating system providing the average rating to a rating distribution system configured to provide the average rating to potential consumers of the content;
- the content rating system associating the rating sources with preference profiles based on consumption habits of the rating sources;
- and
- the content rating system determining a degree of relatedness between a preference profile associated with the rating and preference profiles of each of the rating sources, wherein the weighting of the rating is further based upon a degree of relatedness between the preference profile associated with the rating and the preference profiles associated with the rating sources, and wherein the weighting comprises increasing the weight of the rating if the preference profile of the rating and the preference profiles of the rating sources are determined to have a high degree of relatedness and the weighting comprises decreasing the weight of the rating if the preference profile of the rating and the preference profiles of the rating sources are determined to have a low degree of relatedness.
23 more elements
A tangible, non-transitory content rating computer system for optimizing ratings associated with content, the system comprising:
- a rating adjusting service configured to:
- receive ratings of content from one or more rating sources;
- ascertain whether the rating sources accessed the content for which the ratings were received;
- determine how the ratings provided by each of the rating sources a affect popularity of the content by:
- counting access to the content over a duration of time;
- evaluating a speed of change in the popularity of the content over a duration of time, the speed of popularity change being represented as:
- S=(Pop2−Pop1)/(Time2−Time2),
- where S is the speed of change, Pop2 is a popularity of the content at Time2, and Pop1 is a popularity of the content at Time1;
- and
- evaluating an acceleration of the change in the popularity of the content over the duration of time, the acceleration of popularity change being represented as:
- A=(S2−S1)/(Time2−Time1),
- where A is the acceleration of popularity change, S2 is a speed of popularity change at Time2, and S1 is a speed of popularity change at Time1;
- adjust a weight accorded to the ratings received from each of the rating sources based on the determined and ascertaining, wherein adjusting the weight accorded to the ratings comprises:
- increasing the weight accorded to the ratings if the ratings comprise positive ratings and the popularity of the content increases;
- increasing the weight accorded to the ratings if the ratings comprise negative ratings and the popularity of the content decreases;
- and
- decreasing the weight accorded to the ratings if the ratings comprise negative ratings and the popularity of the content does not decrease;
- decreasing the weight accorded to the ratings if the ratings comprise positive ratings and the popularity of the content does not increase;
- and
- decreasing the weight accorded to the ratings if the rating source did not access the content;
- and
- generate an average rating for the content, the average rating comprising at least two weighted ratings;
- and
- a rating distribution service configured to provide the average rating to potential consumers, wherein the rating adjusting service is further configured to associate the rating sources with preference profiles based on consumption habits of the rating sources and determine a degree of relatedness between a preference profile associated with the rating and preference profiles of each of the rating sources, wherein the weighting of the rating is further based upon a degree of relatedness between the preference profile associated with the rating and the preference profiles associated with the rating sources, and wherein the weighting comprises increasing the weight of the rating if the preference profile of the rating and the preference profiles of the rating sources are determined to have a high degree of relatedness and the weighting comprises decreasing the weight of the rating if the preference profile of the rating and the preference profiles of the rating sources are determined to have a low degree of relatedness.
21 more elements
A tangible, non-transitory content rating computer system for optimizing ratings associated with content, the system comprising:
- a rating adjusting service configured to:
- receive ratings of content from one or more rating sources;
- determine how the ratings provided by each of the rating sources a affect popularity of the content by:
- counting access to the content over a duration of time;
- and
- evaluating a speed of change in the popularity of the content over a duration of time, the speed of popularity change being represented as:
- S=(Pop2−Po1)/(Time2−Time1),
- where S is the speed of change, Pop2 is a popularity of the content at Time2, and Pop1 is a popularity of the content at Time1;
- and
- evaluating an acceleration of the change in the popularity of the content over the duration of time, the acceleration of popularity change being represented as:
- A=(S2−S1)/(Time2−Time1),
- where A is the acceleration of popularity change, S2 is a speed of popularity change at Time2, and S1 is a speed of popularity change at Time1;
- detect whether the content was accessed on multiple occasions by the rating source;
- adjust a weight accorded to the ratings received from each of the rating sources based on the determining and detecting, wherein adjusting the weight accorded to the ratings comprises:
- increasing the weight accorded to the ratings if:
- the ratings comprise positive ratings and the popularity of the content increases;
- or
- the ratings comprise negative ratings and the popularity of the content decreases;
- decreasing the weight accorded to the ratings if:
- the ratings comprise negative ratings and the popularity of the content does not decrease;
- or
- the ratings comprises positive ratings and the popularity of the content does not increase;
- and
- adjusting the weight of the rating based upon whether the content was accessed on multiple occasions by the rating source;
- and
- generate an average rating for the content, the average rating comprising at least two weighted ratings;
- and
- a rating distribution service configured to provide the average rating to potential consumers, wherein the rating adjusting service is further configured to associate the rating sources with preference profiles based on consumption habits of the rating sources and determine a degree of relatedness between a preference profile associated with the rating and preference profiles of each of the rating sources, wherein the weighting of the rating is further based upon a degree of relatedness between the preference profile associated with the rating and the preference profiles associated with the rating sources, and wherein the weighting comprises increasing the weight of the rating if the preference profile of the rating and the preference profiles of the rating sources are determined to have a high degree of relatedness and the weighting comprises decreasing the weight of the rating if the preference profile of the rating and the preference profiles of the rating sources are determined to have a low degree of relatedness.
25 more elements
Dependencies are inferred from the claim text.