Automatic generation of electronic advertising messages containing one or more automatically selected stock photography images

1. A computing system for constructing advertising messages, the computing system comprising:

one or more processors; and

one or more memories storing instructions that, when executed by the one or more processors, cause the computing system to perform a process comprising:

receiving a data feed comprising, for each of multiple textual advertising messages:

an identification of a corresponding web page from which at least one component can be extracted; and

one or more text sections defining features to be included in an advertising message, search parameters for content for the advertising message, parameters for using the advertising message, or any combination thereof; and

for each of the multiple textual advertising messages:

extracting the at least one component from the corresponding web page;

automatically selecting an image for the advertising message by:

determining that the corresponding web page contains an insufficient number of assets a) specified for the advertising message or b) with a specified characteristic; and

in response to the determining the insufficient number of assets, selecting a stock photography image matching the search parameters for content for the advertising message; and

generating the advertising message that:

A) incorporates the automatically selected image and,

B) formats the advertising message consistent with formatting of the corresponding web page,

 wherein the formatting of the generated advertising message is based on the at least one component extracted from the corresponding web page.

2. The computing system of claim 1 , wherein the process further comprises:

causing display of one of the generated advertising messages to a plurality of user,

wherein the causing the display is in accordance with the parameters for using the advertising message in the text section of the textual advertising message corresponding to the one of the generated advertising messages.

3. The computing system of claim 1 , wherein the advertising message is a first advertising message, and wherein generating a second particular advertising message comprises automatically selecting a second image for the second advertising message by:

extracting text from a second corresponding web page identified in a second textual advertising message for the second advertising message; and

executing an image search using a query that is based on one or more terms identified in the extracted text.

4. The computing system of claim 1 , wherein the advertising message is a first advertising message, and wherein generating a second particular advertising message comprises automatically selecting a second image for the second the particular advertising message by:

extracting text from a second corresponding web page identified in a second textual advertising message for the second advertising message;

identifying, for each of multiple subject matter categories, terms in the extracted text that match that subject matter category;

computing a score, for each of the subject matter categories, based upon a number of occurrences of terms in the extracted text identified as matching that subject matter category; and

executing an image search using a query that is based on at least some of the terms matching one of the subject matter categories with the highest score.

5. The computing system of claim 1 , wherein the advertising message is a first advertising message, and wherein generating a second advertising message comprises automatically selecting a second image for the second advertising message by:

extracting text from a second corresponding web page identified in a second textual advertising message for the second advertising message;

identifying, for each of multiple subject matter categories, terms in the extracted text that match that subject matter category;

computing a score, for each of the subject matter categories, based upon:

a number of occurrences of terms in the extracted text identified as matching that subject matter category, and

a weight, assigned to each term in the extracted text identified as matching that subject matter category, indicating the extent to which the term relates to that subject matter category; and

executing an image search using a query that is based on at least some of the terms matching one of the subject matter categories with the highest score.

6. The computing system of claim 1 , wherein the process further comprises:

causing one of the generated advertising messages to be presented to a user in a second web page, other than the corresponding web page from which the at least one component was extracted for that advertising message.

7. The computing system of claim 1 , wherein the advertising message is a first advertising message, and wherein generating a second advertising message comprises automatically selecting a second image for the second advertising message by:

determining that a second corresponding web page contains a reference to a third corresponding web page identified for containing assets that provide for effective display of advertising messages,

wherein the third corresponding web page is identified for containing assets that provide for effective display of advertising messages based on determ ining that the corresponding web_page is a profile page for a business that contains the reference to the profiled business's home page as the third corresponding web page; and

using an image automatically extracted from the third corresponding web page as the automatically selected image when generating the second advertising message.

8. The computing system of claim 1 , wherein the determining that the corresponding web page contains the insufficient number of assets includes determining that a) the corresponding web page contains below the number of assets specified for the particular advertising message.

9. The computing system of claim 1 , wherein the advertising message is a first advertising message, and wherein generating a second advertising message comprises automatically selecting a second image for the the advertising message by:

determining that a second corresponding web page contains below a number of assets with specified quality characteristics for asset format, asset dimensions, or any combination thereof; and

in response to the determining that the second corresponding web page contains below the number of assets with the specified quality characteristics, selecting one or more stock photography images that a) match the search parameters for content for the second advertising message and b) have the specified quality characteristics.

10. The computing system of claim 1 , wherein the process further comprises:

causing display of the advertising message, of at least one of the advertising messages, for a recipient;

wherein placement features, including setting a price or budget for the display and selecting a geographic region of the recipient, used in the display of the advertising message are based on the parameters for using the advertising message in the corresponding textual advertising message.

11. The computing system of claim 1 , wherein the determining that the corresponding web page contains the insufficient number of assets includes determining that the corresponding web page contains below the number of assets with specified characteristics.

12. A computer-readable storage medium storing instructions that, when executed by a computing system, cause the computing system to perform a process comprising:

receiving a data feed comprising, for each of multiple textual sections:

an identification of a corresponding web page from which at least one component can be extracted; and

at least one of: one or more features to be included in a display module, search parameters for content for the display module, parameters for using the display module, or any combination thereof;

automatically selecting and image for the display module by:

determining that the corresponding web page contains a reference to a second corresponding web page identified for containing assets that provide for effective display of display modules by determining that the corresponding web page is a profile page for a business that contains the reference to the profiled business's home page as the second corresponding web page; and

using an image automatically extracted from the second corresponding web page as the automatically selected image when generating the display module;

for each of the multiple textual sections:

extracting the at least one component from the corresponding web page; and

generating the display module that A) incorporates an automatically selected image and B) formats the display module consistent with formatting of the corresponding web page,

wherein the formatting of the generated display module is based on the at least one component extracted from the corresponding web page.

13. The computer-readable storage medium of claim 12 , wherein the process further comprises:

causing display of one of the generated display modules to a plurality of users,

wherein the causing the display is in accordance with the parameters for using the display module that are in the features to be included of the textual section, corresponding to the one of the generated display modules.

14. The computer-readable storage medium of claim 12 , wherein the display module is a first display module, and wherein generating a second display module comprises automatically selecting a second image for the second display module by:

extracting text from a second corresponding web page identified in a second textual section for the second display module; and

executing an image search using a query that is based on one or more terms identified in the extracted text.

15. The computer-readable storage medium of claim 12 , wherein the display module is a first display module, and wherein generating a second display module comprises automatically selecting a second image for the second display module by:

extracting text from a second corresponding web page identified in a second textual section for the second display module;

identifying, for each of multiple subject matter categories, any terms in the extracted text that match that subject matter category;

computing a score, for each of the subject matter categories, based upon a number of occurrences of terms in the extracted text identified as matching that subject matter category; and

executing an image search using a query that is based on at least some of the terms matching one of the subject matter categories with the highest score.

16. The computer-readable storage medium of claim 12 , wherein the display module is a first display module, and wherein generating a second display module comprises automatically selecting a second image for the second display module by:

extracting text from a second corresponding web page identified in a second textual section for the second display module;

identifying, for each of multiple subject matter categories, terms in the extracted text that match that subject matter category;

computing a score, for each of the subject matter categories, based upon:

a number of occurrences of terms in the extracted text identified as matching that subject matter category, and

a weight, assigned to each term in the extracted text identified as matching that subject matter category, indicating the extent to which the term relates to that subject matter category; and

executing an image search using a query that is based on at least some of the terms matching one of the subject matter categories with the highest score.

17. The computer-readable storage medium of claim 12 , wherein the process further comprises:

causing one of the generated display modules to be presented to a user in a second web page, other than the corresponding web page from which the at least one component was extracted for that display module.

18. A method for automatically generating an advertisement, the method comprising:

extracting at least one component from a web page identified in association with the advertisement;

determining that the web page contains below a number of assets with specified quality characteristics for asset format, asset dimensions, or any combination thereof;

in response to the determining that the web page contains below the number of assets with the specified quality characteristics, automatically selecting one or more images that a) match search parameters specified for content for the advertisement and b) have the specified quality characteristics;

generating the advertisement,

incorporating in the advertisement the automatically selected one or more images; and

formatting the advertisement consistent with formatting of the web page,

wherein the formatting of the advertisement is based on the at least one component extracted from the web page.

19. The method of claim 18 , wherein the web page is identified, in association with the advertisement, in a data feed specifying multiple textual sections, each textual section comprising:

an identification of a corresponding web page from which at least one component can be extracted; and

one or more of: a section defining features to be included in the advertisement, search parameters for content for the advertisement, parameters for using the advertisement, or any combination thereof.

20. The method of claim 19 further comprising:

causing display of the advertisement;

wherein placement features, used in the display of the advertisement, include setting a price or budget for the display and selecting a geographic region of a recipient; and

wherein the placement features are selected are based on the parameters for using the advertisement in one of the textual sections corresponding to the advertisement.