Granted Patent B2

US 9,430,565 · App. 14/161,158 · Granted Aug 30, 2016

Inventor: Zacharias Lee James (Venice, CA)

Assignee: ZEFR, INC.

G06F17/30858

Abstract

Claims (50)

1. A method comprising:

indexing, by use of a processor, each content identifier for content received in a search result to content data, wherein each content comprises related content identifiers comprising video content elements comprising a content length, a content description, a content channel, and a search query;

iteratively retrieving content for the indexed content identifiers and indexing the related content identifiers for the retrieved content until all related content identifiers for all the indexed content identifiers are indexed;

calculating a relevance value for each video content element;

calculating a relevance score RV as RV i =Σk i CEF i where k i is a first nonzero constant for a specified content element i and CEF i is one (1) when an instance of a search query element is found in the specified content element i;

identifying relevant content based on the relevance score, wherein the video content is relevant content in response to a relevance score exceeding a relevance threshold;

flagging the relevant content;

determining whether a relevance confirmation for the relevant content exceeds a confirmation threshold, wherein the relevance confirmation is calculated as an engagement score, wherein the engagement score ES is calculated as ES=(a*ACR)+(b*NC) where ACR is an average content rating for the relevant content, NC is a number of comments for the relevant content, and a and b are non-zero constants;

modifying the first nonzero constants for the relevant content in response to the relevance confirmation not exceeding the confirmation threshold;

recalculating the relevance score in response to the relevance confirmation not exceeding the confirmation threshold;

sorting the flagged relevant content based on the relevance score; and

providing the relevant content.

2. The method of claim 1 , wherein previously indexed related content identifiers are not re-indexed.

3. The method of claim 1 , wherein the content is further relevant in response to a content description comprising terms with a specified term density.

4. The method of claim 1 , wherein the content is further relevant in response to a content description comprising terms with a specified proximity to the search query.

5. The method of claim 1 , wherein the content is further relevant in response to a content description being substantially unique within a channel.

6. The method of claim 5 , wherein a specified number of characters of the content description are not evaluated in determining that the content description in substantially unique within the content channel.

7. The method of claim 1 , wherein the content is further relevant in response to a content length being within a specified length range.

8. The method of claim 1 , wherein the content is further relevant in response to content views exceeding a specified view threshold.

9. The method of claim 1 , wherein the content is further relevant in response to a content submission time being after a specified submission time.

10. The method of claim 1 , further comprising sorting the relevant content based on sorting criteria selected from one or more of content submission time, content views, content owner, and claims.

11. The method of claim 1 , further comprising sorting the relevant content based on an engagement score for the content calculated as a function of responses to the relevant content.

12. The method of claim 1 , further comprising receiving a content evaluation for the content comprising a professional/user designation, a formal title, a content type, a target content platform, a target audience, an offensive content type, and a category.

13. An apparatus comprising:

a processor;

a non-transitory memory storing computer readable code executable by the processor to perform:

indexing each content identifier for content received in a search result to content data, wherein each content comprises related content identifiers comprising video content elements comprising a content length, a content description, a content channel, and a search query;

iteratively retrieving content for the indexed content identifiers and indexing the related content identifiers for the retrieved content until all related content identifiers for all the indexed content identifiers are indexed; and

calculating a relevance value for each video content element;

calculation a relevance score RV as RV i =Σk i CEF i where k i is a first nonzero constant for a specified content element i and CEF i is one (1) when an instance of a search query element is found in the specified content element i;

identifying relevant content based on the relevance score, wherein the video content is relevant content is response to a relevance score exceeding a relevance threshold;

flagging the relevant content;

determining whether a relevance confirmation for the relevant content exceeds a confirmation threshold, wherein the relevance confirmation is calculated as a engagement score, wherein the engagement score ES is calculated as ES=(a*ACR)+(b*NC) where ACR is an average content rating for the relevant content, NC is a number of comments for the relevant content, and a and b are non-zero constants;

modifying the first nonzero constants for the relevant content in response to the relevance confirmation not exceeding the confirmation threshold;

recalculating the relevance score in response to the relevance confirmation not exceeding the confirmation threshold;

sorting the flagged relevant content based on the relevance score; and

providing the relevant content.

14. The apparatus of claim 13 , wherein the content is further relevant in response to a content description comprising terms with a specified term density.

15. A program product comprising a non-transitory computer readable storage medium storing program code executable by a processor to perform:

indexing each content identifier for content received in a search result to content data, wherein each content comprises related content identifiers comprising video content elements comprising a content length, a content description, a content channel, and a search query;

iteratively retrieving content for the indexed content identifiers and indexing the related content identifiers for the retrieved content until all related content identifiers for all the indexed content identifiers are indexed;

calculating a relevance value for each video content element;

calculating a relevance score RV as RV i =Σk i CEF i where k i is a first nonzero constant for a specified content element i and CEF i is one (1) when an instance of a search query element is found in the specified content element i;

identifying relevant content based on the relevance score wherein the video content is relevant content in response to a relevance score exceeding a relevance threshold;

flagging the relevant content;

determining whether a relevance confirmation for the relevant content exceeds a confirmation threshold, wherein the relevance confirmation is calculate as an engagement score, wherein the engagement score ES is calculated as ES=(a*ACR)+(b*NC) where ACR is an average content rating for the relevant content, NC is a number of comments for the relevant content, and a and b are non-zero constants;

modifying the first nonzero constants for the relevant content response to the relevance confirmation not exceeding the confirmation threshold;

recalculating the relevance score in response to the relevance confirmation not exceeding the confirmation threshold;

sorting the flagged relevant content based on the relevance score; and

providing the relevant content.

Assignments (16)

Continuity (1)