Selection of advertisements for placement with content

1. A computer-implemented method, the method comprising:

receiving a request for sponsored content for presentation with a multimedia content item, wherein the content item is included in a document;

selecting a plurality of keywords for the content item wherein a first keyword of the keywords is selected based on historical user interaction with the content item when the content item was provided as a search result responsive to a query comprising the first keyword, and wherein a second keyword of the keywords is selected based on analysis of the content item;

for each of one or more candidate sponsored content items:

calculating a respective first score based at least partially on a comparison of the candidate sponsored content item to one or more of the keywords;

calculating a respective second score based on, at least, a comparison of the candidate sponsored content item to content of the document and based on a profile of an owner of the document;

calculating a respective final score based on, at least, a weighted combination of the respective first and second scores in which the first and second scores are weighted differently, and in which a weight of the first or the second score is based at least partially on a measure of past user interactions with the candidate sponsored content item when the candidate sponsored content item was presented with the content item;

selecting one or more of the candidate sponsored content items based on, at least, the candidate sponsored content items' respective final scores; and

providing the selected sponsored content items for presentation with the content item.

2. The method of claim 1 , wherein the sponsored content items comprise one or more advertisements.

3. The method of claim 1 , wherein the content item comprises a video file or stream.

4. The method of claim 1 , wherein calculating one of the scores is further based on a user profile of a user accessing the document or the content item.

5. The method of claim 1 , wherein calculating one of the scores is further based on metadata associated with the content item.

6. The method of claim 1 , wherein calculating one of the scores is further based on content of a second different document that contains a hyperlink to the content item.

7. The method of claim 6 , wherein the content of the second different document that contains the hyperlink to the content item comprises anchor text of the hyperlink.

8. The method of claim 1 , wherein:

the content item includes audio data, and

wherein the analysis of the content item comprises a speech recognition analysis of the audio data.

9. The method of claim 8 , wherein the speech recognition analysis of the audio data comprises a determination of a confidence level for each of one or more terms in the audio data.

10. The method of claim 1 , wherein:

the content item includes visual data; and

wherein the analysis of the content item comprises an image recognition analysis of the visual data.

11. The method of claim 1 , wherein selecting one or more of the candidate sponsored content items comprises selecting one or more of the candidate sponsored content items based on respective bids associated with the candidate sponsored content items.

12. A system comprising:

one or more processors; and

a storage device having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising:

receive a request for sponsored content for presentation with a multimedia content item, wherein the content item is included in a document;

selecting a plurality of keywords for the content item wherein a first keyword of the keywords is selected based on historical user interaction with the content item when the content item was provided as a search result responsive to a query comprising the first keyword, and wherein a second keyword of the keywords is selected based on analysis of the content item;

for each of one or more candidate sponsored content items:

calculate a respective first score based at least partially on a comparison of the candidate sponsored content item to one or more of the keywords;

calculate a respective second score based on, at least, a comparison of the candidate sponsored content item to content of the document and based on a profile of an owner of the document;

calculate a respective final score based on, at least, a weighted combination of the respective first and second scores in which the first and second scores are weighted differently, and in which a weight of the first or the second score is based at least partially on a measure of past user interactions with the candidate sponsored content item when the candidate sponsored content item was presented with the content item;

select one or more of the candidate sponsored content items based on, at least, the candidate sponsored content item's respective final scores; and

provide the selected sponsored content items for presentation with the content item.

13. The system of claim 12 wherein the sponsored content items comprise one or more advertisements.

14. The system of claim 12 wherein the content item comprises a video file or stream.

15. The system of claim 12 wherein calculating one of the scores is further based on a user profile of a user accessing the document or the content item.

16. The system of claim 12 wherein calculating one of the scores is further based on metadata associated with the content item.

17. The system of claim 12 wherein calculating one of the scores is further based on content of a second different document that contains a hyperlink to the content item.

18. The system of claim 17 wherein the content of the second different document that contains the hyperlink to the content item comprises anchor text of the hyperlink.

19. The system of claim 12 wherein:

the content item includes audio data, and

wherein the analysis of the content item comprises a speech recognition analysis of the audio data.

20. The system of claim 19 wherein the speech recognition analysis of the audio data comprises a determination of a confidence level for each of one or more terms in the audio data.

21. The system of claim 12 wherein:

the content item includes visual data; and

wherein the analysis of the content item comprises an image recognition analysis of the visual data.

22. The system of claim 12 wherein selecting one or more of the candidate sponsored content items comprises selecting one or more of the candidate sponsored content items based on respective bids associated with the candidate sponsored content items.

23. One or more disk devices having instructions stored thereon which, when executed by a processor, cause the processor to perform operations comprising:

receiving a request for sponsored content for presentation with a content multimedia item, wherein the content item is included in a document;

selecting a plurality of keywords for the content item wherein a first keyword of the keywords is selected based on historical user interaction with the content item when the content item was provided as a search result responsive to a query comprising the first keyword, and wherein a second keyword of the keywords is selected based on analysis of the content item;

for each of one or more candidate sponsored content items:

calculating a respective first score based at least partially on a comparison of the candidate sponsored content item to one or more of the keywords;

calculating a respective second score based on, at least, a comparison of the candidate sponsored content item to content of the document and based on a profile of an owner of the document;

calculating a respective final score based on, at least, a weighted combination of the respective first and second scores in which the first and second scores are weighted differently, and in which a weight of the first or the second score is based at least partially on a measure of past user interactions with the candidate sponsored content item when the candidate sponsored content item was presented with the content item;

selecting one or more of the candidate sponsored content items based on, at least, the candidate sponsored content item's respective final scores; and

providing the selected sponsored content items for presentation with the content item.

24. The one or more disk devices of claim 23 , wherein the sponsored content items comprise one or more advertisements.

25. The one or more disk devices of claim 23 , wherein the content item comprises a video file or stream.

26. The one or more disk devices of claim 23 , wherein calculating one of the scores is further based on a user profile of a user accessing the document or the content item.

27. The one or more disk devices of claim 23 , wherein calculating one of the scores is further based on metadata associated with the content item.

28. The one or more disk devices of claim 23 , wherein calculating one of the scores is further based on content of a second different document that contains a hyperlink to the content item.

29. The one or more disk devices of claim 28 , wherein the content of the second different document that contains the hyperlink to the content item comprises anchor text of the hyperlink.

30. The one or more disk devices of claim 23 , wherein:

the content item includes audio data, and

wherein the analysis of the content item comprises a speech recognition analysis of the audio data.

31. The one or more disk devices of claim 30 , wherein the speech recognition analysis of the audio data comprises a determination of a confidence level for each of one or more terms in the audio data.

32. The one or more disk devices of claim 23 , wherein:

the content item includes visual data; and

wherein the analysis of the content item comprises an image recognition analysis of the visual data.

33. The one or more disk devices of claim 23 , wherein selecting one or more of the candidate sponsored content items comprises selecting one or more of the candidate sponsored content items based on respective bids associated with the candidate sponsored content items.