Video classification

Patent metadata

Field	Value
Publication number	US-9615136-B1
Application number	US-201313887107-A
Country	US
Kind code	B1
Filing date	May 3, 2013
Priority date	May 3, 2013
Publication date	Apr 4, 2017
Grant date	Apr 4, 2017

Abstract

The present technology may identify item category affinities by identifying a plurality of classifications of an item. An accuracy of the plurality of classifications relative to one another for the item may be identified. A category affinity of the item may be determined based on the accuracy of the plurality of classifications relative to one another.

First claim

The invention claimed is:

1. A method for a network page that represents category affinities for videos, comprising:
generating, using at least one processor, the network page associated with an electronic retailer containing a representation of a video having a classification, and representations of other videos having a same classification;
wherein generating the network page includes:
identifying a plurality of classifications for the video;
determining a ranking of the plurality of classifications relative to one another for the video based on evaluating reviews of the video to identify an occurrence rate of the plurality of classifications within the reviews;
identifying a video browsing activity that includes a search for the video that returns the video and related videos having a classification in common with the video;
increasing the ranking of a classification within the plurality of classifications as a result of identifying a subset of videos having a classification in common with the video that have been selected for viewing or viewing video details as part of the video browsing activity;
wherein, the classification is dissociated from the video when an affinity of the video with the classification is below a predetermined threshold; and
providing the network page to a client device that requested the network page.

2. The method of claim 1 , wherein identifying the ranking comprises identifying a classification common to the video and a second video requested by a client device, and increasing the ranking of the classification common to the video and a second video.

3. The method of claim 1 , further comprising determining whether the video is an accurate representation of the selected classification.

4. The method of claim 3 , wherein determining whether the video is an accurate representation comprises at least one of: receiving ratings of the selected classification or tracking user interaction with respect to the video in the one of the plurality of classifications.

5. A method for generating a network page that represents item category affinities for an item, comprising:
generating by at least one processor the network page, wherein generating the network page includes:
identifying a plurality of classifications of the item;
identifying an accuracy of the plurality of classifications relative to one another for the item;
increasing the accuracy of a classification within the plurality of classifications as a result of identifying a subset of items having a classification in common with the item, the subset of items being included with the item as part of an item browsing activity that includes a search for the item that returns the item and related items having a classification in common with the item and the subset of items selected as part of the item browsing activity;
determining a category affinity of the item based on the accuracy of the plurality of classifications relative to one another, wherein classification is dissociated from the item when an affinity of the item with the classification is below a predetermined threshold; and
providing the network page containing a representation of the item having the category affinity and representations of other items having the category affinity.

6. The method of claim 5 , wherein the accuracy of the plurality of classifications comprises a plurality of estimated percentages indicating a percentage of the item fitting each of the plurality of classifications.

7. The method of claim 6 , wherein a sum of the estimated percentages totals 100%.

8. The method of claim 6 , wherein a sum of the estimated percentages is greater than 100%.

9. The method of claim 6 , wherein a sum of the estimated percentages is less than 100%.

10. The method of claim 5 , further comprising providing a user interface representation of the item for display with representations of other items having a same classification when a request for one of the plurality of classifications is received from a client device.

11. The method of claim 5 , further comprising ranking the item among search results based on account type and request velocity of the item from client devices.

12. The method of claim 5 , further comprising altering a ranking of the item among other items based on a context in which a client device requests the items.

13. The method of claim 5 , further comprising tracking user interaction with the item when a request is received for one of the plurality of classifications, and modifying the accuracy of the one of the plurality of classifications for the item based on the user interaction.

14. The method of claim 5 , further comprising ordering the representation of the video relative to the representations of other videos using an accuracy based on a ranking.

15. The method of claim 5 , wherein the method is implemented as computer readable program code executed by the processor, the computer readable code being embodied on a non-transitory computer usable medium.

16. A system for managing content item category affinities, comprising:
a processor;
a memory device including instructions that, when executed by the processor, cause the system to:
identify a plurality of classifications of a content item;
identify a ranking of the plurality of classifications relative to one another;
identify a content item browsing activity that includes a search for the content item that returns the content item and related content items having a classification in common with the content item;
increase the ranking of a classification within the plurality of classifications as a result of identifying a subset of content items having a classification in common with the content item that have been selected as part of the content item browsing activity; and
dissociate a classification from the content item when an affinity of the content item with the classification is below a predetermined threshold;
generate a network page for an electronic retailer containing a representation of the content item for display on a client device with representations of other content items having a same classification and according to the ranking in response to a request from the client device for the network page containing one of the plurality of classifications.

17. The system of claim 16 , wherein the memory device includes instructions that, when executed by the processor, causes the system to:
analyze a review related to the content item to identify an evaluation attribute from within the review; and
generate a label based on the evaluation attribute and to associate the label with the content item, the label being a classification label.

18. The system of claim 17 , wherein the memory device includes instructions that, when executed by the processor, causes the system to receive a search query from a user and generate search results in response to the search query and based on the label.

19. The system of claim 18 , wherein the memory device includes instructions that, when executed by the processor, causes the system to provide a cluster of content items having the label associated as the search results for display on a client device.

20. The system of claim 17 , wherein the memory device includes instructions that, when executed by the processor, causes the system to cluster personalized and non-personalized content items together for display on the client device, the personalized content items being based on a history or preference stored with an account associated with the c

Assignees

• Amazon Tech Inc

Inventors

• Emery Grant Michael
• Bhagat Rahul Hemant
• Cameros Brian
• Cohen Benjamin Thomas
• Dillard Logan Luyet
• Liang Yongwen
• Mongrain Scott Allen
• Quinn Michael David
• Rosofsky Eli Glen
• Sanders Adam Callahan

Classifications

• G06F16/64

  Browsing; Visualisation therefor (generation of a list or set of audio data G06F16/638) · CPC title

• H04N21/4756

  for rating content, e.g. scoring a recommended movie · CPC title

• G06F16/743

  a collection of video files or sequences · CPC title

• H04N21/25891

  being end-user preferences (retrieval of video data in a video database based on user preferences G06F16/739; arrangements for recognizing users' preferences H04H60/46; user profiles in network data switching protocols H04L67/306; processing of user preferences or user profiles in wireless networks H04W8/18) · CPC title

• G06F16/44

  Browsing; Visualisation therefor · CPC title