Analyzing website content or attributes and predicting popularity

• US 9,122,989 B1
• Filed: 03/15/2013
• Issued: 09/01/2015
• Est. Priority Date: 01/28/2013
• Status: Active Grant

• Current Assignee: XANT Inc.
• Original Assignee: Insidesales.com
• Inventors: Zeng, Xinchuan, Morris, Richard, Elkington, David, Krogue, Kenneth
• Primary Examiner(s): CHEN, ALAN S
• Application Number:
  US13/840,424
• Time in Patent Office: 900 Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

A method to analyze and determine which source content and user interactions are most popular is provided. The method generates scores for items, e.g., articles, topics, authors, or influencers, on a particular source based on data gathered from both the particular source and social media sources. The scores are used to rank items of the same type, and determine which items are the most popular. The method may also take demographic information as input. Using the demographic information, the system may determine the popularity of a particular item in a particular demographic. The method may also predict which demographic an item may be the most popular in. Furthermore, the method may give a recommendation on which author should write on a particular topic, which topic is most likely to be the most popular for a particular demographic, and which influencers should promote the article.

Claims

1. A method comprising:

computing a first score for a first item that indicates popularity of the first item across all demographics;

computing a second score for a second item that indicates popularity of the second item across all demographics;

wherein the first item and the second item are a same item type;

computing a first impact score that indicates how popular the first item is for a particular demographic relative to other demographics;

computing a second impact score that indicates how popular the second item is for the particular demographic relative to other demographics;

computing a first demographic-specific popularity score for the first item based, at least in part, on the first score and the first impact score;

computing a second demographic-specific popularity score for the second item based, at least in part, on the second score and the second impact score; and

predicting that the particular demographic has a higher interest in the first item than in the second item based on the first demographic-specific popularity score being greater than the second demographic-specific popularity score;

wherein the method is performed by one or more computing devices.

2. The method of claim 1, wherein the same item type is one of articles, topics, authors, or influencers.

3. The method of claim 1 comprising:

selecting the particular demographic by selecting attributes, from a plurality of available attributes, that are possessed by a particular lead;

predicting that the particular lead has the higher interest in the first item than in the second item based on the first demographic-specific popularity score being greater than the second demographic-specific popularity score.

4. The method of claim 1, wherein:

the first item is a first article and the second item is a second article; and

further comprising predicting that people in the particular demographic are more likely to have read the first article based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

5. The method of claim 1, wherein the first item is a first topic and the second item is a second topic, and further comprising:

predicting that the first topic is more relevant to the particular demographic based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

6. The method of claim 1, wherein:

the first item is a first author and the second item is a second author; and

the method further comprises predicting that an article written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

7. The method of claim 1, wherein:

the first item is a first influencer and the second item is a second influencer, and

the method further comprises predicting that an article promoted by the first influencer is more likely to attract more readers in the particular demographic than the article promoted by the second influencer based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

8. The method of claim 1, wherein:

the first item is a first author and the second item is a second author;

more weight is given to factors that relate to a particular topic, when computing the first score for the first item, than to factors that relate to topics other than the particular topic;

more weight is given to factors that relate to the particular topic, when computing the second score for the second item, than to factors that relate to topics other than the particular topic; and

the method further comprises predicting that an article related to the particular topic and written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

9. The method of claim 1, wherein:

the first item is a first author and the second item is a second author; and

the method further comprises predicting that an article on a particular topic and written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the second author has not written any articles on the particular topic.

10. The method of claim 1, wherein:

the first item is a first influencer and the second item is a second influencer;

more weight is given to factors that relate to a particular topic, when computing the first score for the first item, than is given to factors that relate to topics other than the particular topic;

more weight is given to factors that relate to the particular topic, when computing the second score for the second item, than is given to factors that relate to topics other than the particular topic; and

the method further comprises predicting that the first influencer is more likely than the second influencer to attract readers in the particular demographic to an article related to the particular topic based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

11. The method of claim 1, wherein:

the first item is a first influencer and the second item is a second influencer; and

the method further comprises predicting the first influencer is more likely than the second influencer to attract readers in the particular demographic to an article on a particular topic based, at least in part, on determining that the second influencer has not promoted any article on the particular topic.

12. A non-transitory computer-readable medium storing instructions which, when executed by one or more processors, cause performance of a method comprising:

computing a first score for a first item that indicates popularity of the first item across all demographics;

computing a second score for a second item that indicates popularity of the second item across all demographics;

wherein the first item and the second item are a same item type;

wherein the same item type is one of articles, topics, authors, or influencers;

computing a first impact score that indicates how popular the first item is for a particular demographic relative to other demographics;

computing a second impact score that indicates how popular the second item is for the particular demographic relative to other demographics;

computing a first demographic-specific popularity score for the first item based, at least in part, on the first score and the first impact score;

computing a second demographic-specific popularity score for the second item based, at least in part, on the second score and the second impact score; and

predicting that the particular demographic has a higher interest in the first item than in the second item based on the first demographic-specific popularity score being greater than the second demographic-specific popularity score.

13. The non-transitory computer-readable medium of claim 12 comprising:

selecting the particular demographic by selecting attributes, from a plurality of available attributes, that are possessed by a particular lead;

predicting that the particular lead has the higher interest in the first item than in the second item based on the first demographic-specific popularity score being greater than the second demographic-specific popularity score.

14. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first article and the second item is a second article; and

the method further comprising predicting that people in the particular demographic are more likely to have read the first article based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

15. The non-transitory computer-readable medium of claim 12, wherein the first item is a first topic and the second item is a second topic, and the method further comprising:

predicting that the first topic is more relevant to the particular demographic based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

16. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first author and the second item is a second author; and

the method further comprises predicting that an article written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

17. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first influencer and the second item is a second influencer, and

the method further comprises predicting that an article promoted by the first influencer is more likely to attract more readers in the particular demographic than the article promoted by the second influencer based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

18. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first author and the second item is a second author;

more weight is given to factors that relate to a particular topic, when computing the first score for the first item, than to factors that relate to topics other than the particular topic;

more weight is given to factors that relate to the particular topic, when computing the second score for the second item, than to factors that relate to topics other than the particular topic; and

the method further comprises predicting that an article related to the particular topic and written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

19. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first author and the second item is a second author; and

the method further comprises predicting that an article on a particular topic and written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the second author has not written any articles on the particular topic.

20. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first influencer and the second item is a second influencer;

more weight is given to factors that relate to a particular topic, when computing the first score for the first item, than is given to factors that relate to topics other than the particular topic;

more weight is given to factors that relate to the particular topic, when computing the second score for the second item, than is given to factors that relate to topics other than the particular topic; and

the method further comprises predicting that the first influencer is more likely than the second influencer to attract readers in the particular demographic to an article related to the particular topic based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

21. The non-transitory computer-readable medium of claim 12, wherein:

the first item is a first influencer and the second item is a second influencer; and

the method further comprises predicting the first influencer is more likely than the second influencer to attract readers in the particular demographic to an article on a particular topic based, at least in part, on determining that the second influencer has not promoted any article on the particular topic.

22. A computer system comprising:

a memory;

one or more processors coupled to the memory and configured to;

compute a first score for a first item that indicates popularity of the first item across all demographics;

compute a second score for a second item that indicates popularity of the second item across all demographics;

wherein the first item and the second item are a same item type;

compute a first impact score that indicates how popular the first item is for a particular demographic relative to other demographics;

compute a second impact score that indicates how popular the second item is for the particular demographic relative to other demographics;

compute a first demographic-specific popularity score for the first item based, at least in part, on the first score and the first impact score;

compute a second demographic-specific popularity score for the second item based, at least in part, on the second score and the second impact score; and

predict that the particular demographic has a higher interest in the first item than in the second item based on the first demographic-specific popularity score being greater than the second demographic-specific popularity score.

23. The computer system of claim 22, wherein the same item type is one of articles, topics, authors, or influencers.

24. The computer system of claim 22, wherein the one or more processors are further configured to:

select the particular demographic by selecting attributes, from a plurality of available attributes, that are possessed by a particular lead;

predict that the particular lead has the higher interest in the first item than in the second item based on the first demographic-specific popularity score being greater than the second demographic-specific popularity score.

25. The computer system of claim 22, wherein:

the first item is a first article and the second item is a second article; and

the one or more processors are further configured to predict that people in the particular demographic are more likely to have read the first article based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

26. The computer system of claim 22, wherein:

the first item is a first author and the second item is a second author; and

the one or more processors are further configured to predict that an article written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

27. The computer system of claim 22, wherein:

the first item is a first author and the second item is a second author;

more weight is given to factors that relate to a particular topic, when computing the first score for the first item, than to factors that relate to topics other than the particular topic;

more weight is given to factors that relate to the particular topic, when computing the second score for the second item, than to factors that relate to topics other than the particular topic; and

the one or more processors are further configured to predict that an article related to the particular topic and written by the first author is more likely to attract readers in the particular demographic than the second author based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.

28. The computer system of claim 22, wherein:

the first item is a first influencer and the second item is a second influencer;

more weight is given to factors that relate to a particular topic, when computing the first score for the first item, than is given to factors that relate to topics other than the particular topic;

more weight is given to factors that relate to the particular topic, when computing the second score for the second item, than is given to factors that relate to topics other than the particular topic; and

the one or more processors are further configured to predict that the first influencer is more likely than the second influencer to attract readers in the particular demographic to an article related to the particular topic based, at least in part, on determining that the first demographic-specific popularity score is greater than the second demographic-specific popularity score.