Filtering machine for sponsored content

Patent metadata

Field	Value
Publication number	US-2017193545-A1
Application number	US-201514984853-A
Country	US
Kind code	A1
Filing date	Dec 30, 2015
Priority date	Dec 30, 2015
Publication date	Jul 6, 2017
Grant date	—

Abstract

A filtering machine receives sponsored content and filters the sponsored content according to a quality metric generated by quality model circuitry and assigned to the instance of sponsored content. The quality model circuitry generates the quality metric in accordance with historical feedback received about other sponsored content and a collection of quality factors pertaining to the sponsored content. Based on the quality metric for the sponsored content, the filtering machine can effect service of the sponsored content to a user device for display thereon.

First claim

1 . A method comprising:
receiving, at filtering circuitry via communication circuitry, first sponsored content from a first client device; inputting, by the filtering circuitry, at least a portion of the first sponsored content into quality model circuitry; generating, by the quality model circuitry, a quality metric for the first sponsored content according to historical feedback associated with other sponsored content; comparing, by the filtering circuitry, the quality metric to a quality threshold; and effecting provision of the first sponsored content to a second client device for display on the second client device when the quality metric exceeds the quality threshold.

2 . The method of claim 1 further comprising:
implementing, by the quality model circuitry, a predictive model to generate the quality metric based on a plurality of quality factors of the first sponsored content.

3 . The method of claim 2 wherein the plurality of quality factors comprises at least one of an aesthetic appeal factor of the first sponsored content, a trustworthiness factor of the first sponsored content, a brand factor of the first sponsored content, a product/service factor of first sponsored content, a clarity factor of the first sponsored content, or a layout factor of the first sponsored content.

4 . The method of claim 2 wherein implementing the predictive model further comprises implementing a logistic regression predictive model.

5 . The method of claim 2 further comprising:
receiving, by feedback circuitry, feedback associated with second sponsored content; and
training, by the quality model circuitry, a coefficient of the predictive model with the second sponsored content.

6 . The method of claim 1 further comprising:
providing an indication to the first client device that the first sponsored content is poor quality when the quality metric does not exceed the quality threshold.

7 . The method of claim 1 wherein receiving the first sponsored content further comprises receiving an image to be displayed and text to be displayed with the image, and wherein the method further comprises coupling, by intake circuitry, information about the subject matter of the first sponsored content with the first sponsored content.

8 . The method of claim 1 wherein the first sponsored content comprises an advertisement and wherein the other sponsored content comprises another advertisement, and wherein the historical feedback associated with other sponsored content comprises user feedback that the other advertisement is offensive.

9 . The method of claim 1 wherein effecting provision of the first sponsored content to a second client device for display on the second client device further comprises a sponsored content serving circuitry serving the first sponsored content to the second client device.

10 . A machine comprising:
communication circuitry coupled to at least one network and configured to receive first sponsored content from a first client device; quality model circuitry; and filtering circuitry coupled to the communication circuitry and to the quality model circuitry, the filtering circuitry configured to:
receive the first sponsored content from the communication circuitry;
input at least a portion of the first sponsored content into the quality model circuitry;
receive from the quality model circuitry a quality metric for the first sponsored content;
compare the quality metric to a quality threshold; and
effect provision of the first sponsored content to a second client device for display on the second client device when the quality metric exceeds the quality threshold;
wherein the quality model circuitry is configured to generate the quality metric for the first sponsored content according to historical feedback associated with other sponsored content.

11 . The machine of claim 10 wherein the quality model circuitry is further configured to implement a predictive model to generate the quality metric based on a plurality of quality factors of the first sponsored content.

12 . The machine of claim 11 wherein the quality model circuitry is further configured to implement the predictive model to generate the quality metric based on at least one of an aesthetic appeal factor of the first sponsored content, a trustworthiness factor of the first sponsored content, a brand factor of the first sponsored content, a product/service factor of the first sponsored content, a clarity factor of the first sponsored content, or a layout factor of the first sponsored content.

13 . The machine of claim 11 wherein the quality model circuitry is further configured to implement a logistic regression predictive model.

14 . The machine of claim 11 further comprising:
feedback circuitry coupled to the communication circuitry and the quality model circuitry, wherein the feedback circuitry is configured to receive feedback associated with second sponsored content; and
wherein the quality model circuitry is further configured to train a coefficient of the predictive model with the second sponsored content.

15 . The machine of claim 10 wherein the filtering circuitry is further configured to provide an indication to the first client device that the first sponsored content is poor quality when the quality metric does not exceed the quality threshold.

16 . The machine of claim 10 wherein the first sponsored content further comprises an image to be displayed and text to be displayed with the image, and wherein information about the subject matter of the first sponsored content is coupled with the first sponsored content.

17 . The machine of claim 10 wherein the communication circuitry is configured to receive first sponsored content in the form of an advertisement from a first client device and wherein the quality model circuitry is configured to generate the quality metric for the advertisement based on another advertisement and user feedback that the other advertisement is offensive.

18 . A method comprising:
receiving, at filtering circuitry via communication circuitry, first sponsored content from a first client device and second sponsored content from a third client device; inputting, by the filtering circuitry, at least a portion of the first sponsored content and at least a portion of the second sponsored content into quality model circuitry; implementing, by the quality model circuitry, a predictive model; generating, by the quality model circuitry, a first quality metric for the first sponsored content according to historical feedback associated with other sponsored content and based on a plurality of quality factors of the first sponsored content; generating, by the quality model circuitry, a second quality metric for the second sponsored content according to the historical feedback associated with the other sponsored content and based on a plurality of quality factors of the second sponsored content; comparing, by the filtering circuitry, the first quality metric and the second quality metric; and effecting provision of the first sponsored content to a second client device for display on the second client device with higher priority than the second sponsored content when the first quality metric of the first sponsored content exceeds the second quality metric of the second sponsored content.

19 . The method of 18 further comprising:
receiving, by feedback circuitry, negative feedback associated with third sponsored content; and
training, by the quality model circuitry, a coefficient of the predictive

Assignees

• Yahoo Inc

Inventors

• Zhou Ke
• Redi Miriam
• Lalmas Mounia
• Sangal Puneet Mohan

Classifications

• G06Q30/0243 Primary

  Comparative campaigns · CPC title

• G06Q30/0254

  based on statistics · CPC title

• G06Q30/0242

  Determining effectiveness of advertisements · CPC title

• G06Q30/0251

  Targeted advertisements · CPC title

• G06Q30/0244

  Optimization · CPC title