Temporal content selection

1. A method for temporal content selection via a computer network, comprising:

parsing, via at least one processor executing on a server, a first plurality of online documents to generate a first data structure comprising historical keyword co-occurrences;

generating, via the at least one processor parsing a second plurality of online documents different from the first plurality of online documents, a second data structure comprising keyword co-occurrences not in the first data structure;

temporally linking a first keyword of the second data structure with a second keyword of the second data structure, the second keyword co-occurring with the first keyword on a subset of the second plurality of online documents; and

transmitting, via a communication interface of the server, the temporally linked second keyword to a content selection server.

2. The method of claim 1 , wherein parsing the first plurality of online documents comprises:

parsing at least one of an online encyclopedia and an online dictionary.

3. The method of claim 1 , comprising:

identifying a third plurality of online documents via an online news stream, each of the third plurality of online documents comprising a publish time stamp;

comparing the publish time stamp of each of the third plurality of online documents with a temporal threshold to identify the second plurality of online documents.

4. The method of claim 1 , comprising:

evaluating the subset of the second plurality of online documents on which the first keyword and the second keyword co-occur to determine a characteristic of the co-occurrence of the first keyword and the second keyword; and

temporally linking the first keyword with the second keyword responsive to the characteristic satisfying a predetermined threshold for the characteristic.

5. The method of claim 1 , comprising:

determining a frequency of co-occurrence of the first keyword and the second keyword based on the subset of the second plurality of online documents on which the first keyword and the second keyword co-occur; and

temporally linking the first keyword with the second keyword responsive to the frequency of co-occurrence exceeding a frequency of co-occurrence threshold.

6. The method of claim 1 , comprising:

assigning an attribute to the temporal link.

7. The method of claim 1 , comprising:

evaluating at least one of the publish time stamp of the subset of the second plurality of online documents and a frequency of co-occurrence of the first keyword and the second keyword to generate an attribute of the temporal link; and

assigning the attribute to the temporal link.

8. The method of claim 1 , further comprising:

evaluating the subset of the plurality of online documents to determine a duration attribute for the temporal link; and

assigning the duration attribute to the temporal link.

9. The method of claim 8 , further comprising:

removing the temporal link responsive to a duration of the temporal link exceeding the assigned duration attribute.

10. The method of claim 1 , further comprising:

receiving, by the server, an indication from a content provider to select content of the content provider based on temporal links.

11. A system for temporal content selection via a computer network, comprising:

a server comprising one or more processors and memory;

an online site aggregator executing on the server parsing a first plurality of online documents to generate a first data structure comprising historical keyword co-occurrences;

the online site aggregator parsing a second plurality of online documents different from the first plurality of online documents to generate a second data structure comprising keyword co-occurrences not in the first data structure;

a keyword linker, executing on the server, temporally linking a first keyword of the second data structure with a second keyword of the second data structure, the second keyword co-occurring with the first keyword on a subset of the second plurality of online documents; and

a communication interface of the server transmitting the temporally linked second keyword to a content selection server.

12. The system of claim 11 , wherein the first plurality of online documents comprises at least one of an online encyclopedia and an online dictionary.

13. The system of claim 11 , wherein:

the online site aggregator identifies a third plurality of online documents via an online news stream, each of the third plurality of online documents comprising a publish time stamp; and

the online site aggregator compares the publish time stamp of each of the third plurality of online documents with a temporal threshold to identify the second plurality of online documents.

14. The system of claim 11 , wherein the keyword linker:

determines a characteristic of the co-occurrence of the first keyword and the second keyword related to the subset of the second plurality of online documents on which the first keyword and the second keyword co-occur; and

temporally links the first keyword with the second keyword responsive to the characteristic satisfying a predetermined threshold for the characteristic.

15. The system of claim 11 , wherein the keyword linker:

determines a frequency of co-occurrence of the first keyword and the second keyword based on the subset of the second plurality of online documents on which the first keyword and the second keyword co-occur; and

temporally links the first keyword with the second keyword responsive to the frequency of co-occurrence exceeding a frequency of co-occurrence threshold.

16. The system of claim 11 , wherein the keyword linker assigns an attribute to the temporal link.

17. The system of claim 11 , wherein the keyword linker:

evaluates at least one of the publish time stamp of the subset of the second plurality of online documents and a frequency of co-occurrence of the first keyword and the second keyword to generate an attribute of the temporal link; and

assigns the attribute to the temporal link.

18. The system of claim 11 , wherein the keyword linker:

evaluating the subset of the plurality of online documents to determine a duration attribute for the temporal link;

assigns the duration attribute to the temporal link; and

removes the temporal link responsive to a duration of the temporal link exceeding the assigned duration attribute.

19. A non-transitory computer-readable storage medium comprising instructions for temporal content selection via a computer network, the instructions comprising instructions to:

parse a first plurality of online documents to generate a first data structure comprising historical keyword co-occurrences;

parse a second plurality of online documents different from the first plurality of online documents to generate a second data structure comprising keyword co-occurrences not in the first data structure;

temporally link a first keyword of the second data structure with a second keyword of the second data structure, the second keyword co-occurring with the first keyword on a subset of the second plurality of online documents; and

transmit the temporally linked second keyword to a content selection server.

20. The computer-readable storage medium of claim 19 , wherein the instructions further comprise instructions to:

evaluate the subset of the plurality of online documents to determine a duration attribute for the temporal link;

assign the duration attribute to the temporal link; and

remove the temporal link responsive to the duration of the temporal link exceeding the assigned duration attribute.