Content item selection based on presentation context

1. A computer-implemented method, comprising:

identifying a given content item that is eligible to be transmitted to a user device; obtaining, for the given content item, a predicted performance measure;

determining, by one or more computers, that the given content item will be presented at the user device in a navigational context, including determining that a similarity between a search query and a domain name of a landing page for the given content item meets a navigational similarity threshold;

obtaining a navigational performance adjustment factor that indicates a relative performance of content items presented in the navigational context relative to performance of content items presented in a non-navigational context, wherein the non-navigational context is a context in which the search query and the domain name of the content items does not meet the navigational similarity threshold;

adjusting, by the one or more computers and using the navigational performance adjustment factor for a content item slot in which the given content item will be presented, the predicted performance measure of the given content item based on the determination that the given content item will be presented in the navigational context;

selecting, based on the adjusted predicted performance measure, the given content item from among a plurality of available content items; and

updating a user interface of the user device to present the given content item based on the selection of the given content item.

2. The method of claim 1 , further comprising:

receiving a request for a content item from the user device, the request specifying one of the plurality of content item slots of the resource and a query associated with the request; and

obtaining a plurality of eligible content items that are eligible to be provided in response to the request.

3. The method of claim 2 , further comprising:

obtaining a navigational score for the given content item, the navigational score indicating a measure of likelihood that the given content item is navigational for the query associated with the request, wherein:

determining that the given content item will be presented at the user device in a navigational context comprises:

determining that the navigational score for the given content item meets a navigational score threshold; and

in response to determining that the navigational score for the given content item meets the navigational score threshold, determining that the given content item is to be presented in the navigational context.

4. The method of claim 1 , wherein adjusting the predicted performance measure of the given content item comprises increasing the predicted performance measure of the given content item by multiplying the predicted performance measure of the given content item by the navigational performance adjustment factor that corresponds to a content item slot specified by the request.

5. The method of claim 1 , further comprising determining, from first performance data and second performance data, the navigational performance adjustment factor for the content item slot, the navigational performance adjustment factor indicating, for the content item slot, a performance of content items presented in the navigational context relative to a performance of content items presented in the non-navigational context, wherein the first performance data represents performance of content items presented in the navigational context relative and the second performance data represents performance of content items presented in the non-navigational context.

6. The method of claim 5 , wherein:

the first performance data comprises a first click-through rate;

the second performance data comprises a second click-through rate; and

the predicted performance measure is a predicted click-through rate.

7. A system comprising:

a data processing apparatus; and

a data store storing instructions that, when executed by the data processing apparatus, cause the data processing apparatus to perform operations comprising:

identifying a given content item that is eligible to be transmitted to a user device;

obtaining, for the given content item, a predicted performance measure;

determining that the given content item will be presented at the user device in a navigational context, including determining that a similarity between a search query and a domain name of a landing page for the given content item meets a navigational similarity threshold;

obtaining a navigational performance adjustment factor that indicates a relative performance of content items presented in the navigational context relative to performance of content items presented in a non-navigational context, wherein the non-navigational context is a context in which the search query and the domain name of the content items does not meet the navigational similarity threshold;

adjusting, using the navigational performance adjustment factor for a content item slot in which the given content item will be presented, the predicted performance measure of the given content item based on the determination that the given content item will be presented in the navigational context;

selecting, based on the adjusted predicted performance measure, the given content item from among a plurality of available content items; and

updating a user interface of the user device to present the given content item based on the selection of the given content item.

8. The system of claim 7 , wherein the instructions cause the data processing apparatus to perform operations further comprising:

receiving a request for a content item from the user device, the request specifying one of the plurality of content item slots of the resource and a query associated with the request; and

obtaining a plurality of eligible content items that are eligible to be provided in response to the request.

9. The system of claim 8 , wherein the instructions cause the data processing apparatus to perform operations further comprising:

obtaining a navigational score for the given content item, the navigational score indicating a measure of likelihood that the given content item is navigational for the query associated with the request, wherein:

determining that the given content item will be presented at the user device in a navigational context comprises:

determining that the navigational score for the given content item meets a navigational score threshold; and

in response to determining that the navigational score for the given content item meets the navigational score threshold, determining that the given content item is to be presented in the navigational context.

10. The system of claim 7 , wherein adjusting the predicted performance measure of the given content item comprises increasing the predicted performance measure of the given content item by multiplying the predicted performance measure of the given content item by the navigational performance adjustment factor that corresponds to a content item slot specified by the request.

11. The method of claim 7 , wherein the instructions cause the data processing apparatus to perform operations further comprising determining, from first performance data and second performance data, the navigational performance adjustment factor for the content item slot, the navigational performance adjustment factor indicating, for the content item slot, a performance of content items presented in the navigational context relative to a performance of content items presented in the non-navigational context, wherein the first performance data represents performance of content items presented in the navigational context relative and the second performance data represents performance of content items presented in the non-navigational context.

12. The system of claim 11 , wherein:

the first performance data comprises a first click-through rate;

the second performance data comprises a second click-through rate; and

the predicted performance measure is a predicted click-through rate.

13. A non-transitory computer readable medium comprising instructions that, when executed by a data processing apparatus, cause the data processing apparatus to perform operations comprising:

identifying a given content item that is eligible to be transmitted to a user device;

obtaining, for the given content item, a predicted performance measure;

determining that the given content item will be presented at the user device in a navigational context, including determining that a similarity between a search query and a domain name of a landing page for the given content item meets a navigational similarity threshold;

obtaining a navigational performance adjustment factor that indicates a relative performance of content items presented in the navigational context relative to performance of content items presented in a non-navigational context, wherein the non-navigational context is a context in which the search query and the domain name of the content items does not meet the navigational similarity threshold;

adjusting, using the navigational performance adjustment factor for a content item slot in which the given content item will be presented, the predicted performance measure of the given content item based on the determination that the given content item will be presented in the navigational context;

selecting, based on the adjusted predicted performance measure, the given content item from among a plurality of available content items; and

updating a user interface of the user device to present the given content item based on the selection of the given content item.

14. The non-transitory computer readable medium of claim 13 , wherein the instructions cause the data processing apparatus to perform operations further comprising:

receiving a request for a content item from the user device, the request specifying one of the plurality of content item slots of the resource and a query associated with the request; and

obtaining a plurality of eligible content items that are eligible to be provided in response to the request.

15. The non-transitory computer readable medium of claim 14 , wherein the instructions cause the data processing apparatus to perform operations further comprising:

obtaining a navigational score for the given content item, the navigational score indicating a measure of likelihood that the given content item is navigational for the query associated with the request, wherein:

determining that the given content item will be presented at the user device in a navigational context comprises:

determining that the navigational score for the given content item meets a navigational score threshold; and

in response to determining that the navigational score for the given content item meets the navigational score threshold, determining that the given content item is to be presented in the navigational context.

16. The non-transitory computer readable medium of claim 13 , wherein adjusting the predicted performance measure of the given content item comprises increasing the predicted performance measure of the given content item by multiplying the predicted performance measure of the given content item by the navigational performance adjustment factor that corresponds to a content item slot specified by the request.

17. The non-transitory computer readable medium of claim 13 , wherein the instructions cause the data processing apparatus to perform operations further comprising determining, from first performance data and second performance data, the navigational performance adjustment factor for the content item slot, the navigational performance adjustment factor indicating, for the content item slot, a performance of content items presented in the navigational context relative to a performance of content items presented in the non-navigational context, wherein the first performance data represents performance of content items presented in the navigational context relative and the second performance data represents performance of content items presented in the non-navigational context.

18. The non-transitory computer readable medium of claim 11 , wherein:

the first performance data comprises a first click-through rate;

the second performance data comprises a second click-through rate; and

the predicted performance measure is a predicted click-through rate.