Publish-subscribe based methods and apparatuses for associating data files
Various methods and apparatuses are provided which may be implemented using one or more computing devices within a networked computing environment to employ publish-subscribe techniques to associate subscriber encoded data files with a set of publisher encoded data files.
1. A method comprising: with a computing platform,
generating and maintaining a subscriber index for one or more of a plurality of subscriber encoded content files comprising at least informational story content, and a publisher index for one or more of a plurality of publisher encoded content files comprising at least micro-blog and/or social network content;
determining a content-based score and a recency score for the one or more of the plurality of publisher encoded content files for association with the one or more of the plurality of subscriber encoded content files, and using, at least in part, the determined content-based score and the determined recency score to determine a ranking score for the one or more of the plurality of publisher encoded content files for the association with the one or more of the plurality of subscriber encoded content files;
determining for the one or more of the plurality of subscriber encoded content files a top-k set of publisher encoded content files based, at least in part, on the determined ranking score;
in response to obtaining a new publisher encoded content file:
querying the subscriber index using at least a portion of the new publisher encoded content file to determine a ranking score for the new publisher encoded content file with regard to the one or more of the plurality of subscriber encoded content files, wherein the ranking score is based, at least in part, on a content-based score and a recency score, wherein the recency score is based at least in part on a difference between a time of page view and a time of creation of the new publisher encoded content file; and
determining whether the new publisher encoded content file replaces one publisher encoded content file in the top-k set of publisher encoded content files based, at least in part, on a comparison of the ranking score of the new publisher encoded content file and the ranking score of the top-k set of publisher encoded content files.
2. The method as recited in claim 1 , wherein the recency score decreases over a time-interval from the time of creation to the time of page view, and the new publisher encoded content file is included in the top-k set of publisher encoded content files providing the ranking score exceeds a threshold ranking score associated with the top-k set of publisher encoded content files.
3. The method as recited in claim 1 , and further comprising, with the computing platform: in response to a request for the at least one of the plurality of subscriber encoded content files, identifying the top-k set of publisher encoded content files.
4. The method as recited in claim 1 , wherein determining the top-k set of publisher encoded content files further comprises: ranking at least the publisher encoded content files.
5. The method as recited in claim 1 , wherein determining the top-k set of publisher encoded content files further comprises:
determining the top-k set of publisher encoded content files using a top-k retrieval publish-subscribe process comprising at least one of: a term-at-a-time (TAAT) publish-subscribe process; a skipping TAAT publish-subscribe process; a document-at-a-time (DAAT) publish-subscribe process; or a skipping DAAT publish-subscribe process.
6. The method of claim 1 , wherein the indices are generated at least in part based on an indication of content relevancy.
7. The method of claim 1 , wherein replacing one publisher encoded content file in the top-k set of publisher encoded content files comprises: comparing the ranking score of the new publisher encoded content file with the ranking score of one or more publisher encoded content files of the top-k set of publisher encoded content files, and responsive to the comparison, removing a publisher encoded content file with a lower ranking score from the top-k set of publisher encoded content files and adding the new publisher encoded content file to the top-k set of publisher encoded content files.
8. The method of claim 7 , further comprising: receiving a new subscriber encoded content file; querying the publisher index based, at least in part, on a portion of the new subscriber encoded content file; and retrieving a set of publisher encoded content files responsive to the query.
9. A computing platform comprising:
memory; and
a processing unit to:
generate and maintain, in the memory, a subscriber index for one or more of a plurality of subscriber encoded content files to comprise at least informational story content, and a publisher index for one or more of a plurality of publisher encoded content files to comprise at least micro-blog and/or social network content;
determine a content-based score and a recency score for the one or more of the plurality of publisher encoded content files for association with the one or more of the plurality of subscriber encoded content files, and use, at least in part, the to be determined content-based score and the to be determined recency score to determine a ranking score for the one or more of the plurality of publisher encoded content files for the association with the one or more of the plurality of subscriber encoded content files;
determine for the one or more of the plurality of subscriber encoded content files a top-k set of publisher encoded content files to be based at least in part on the to be determined ranking score;
in response to a new publisher encoded content file:
query the subscriber index, at least a portion of the new publisher encoded content file to be employed to determine a ranking score for the new publisher encoded content file with regard to the one or more of the plurality of subscriber encoded content files, wherein the ranking score is to be based, at least in part, on a content-based score and a recency score, wherein the recency score is to be based at least in part on a difference between a time of page view and a time of creation of the new publisher encoded content file; and
determine whether the new publisher encoded content file is to replace one publisher encoded content file in the top-k set of publisher encoded content files to be based, at least in part, on a comparison of the ranking score of the new publisher encoded content file and the ranking score of the top-k set of publisher encoded content files.
10. The computing platform as recited in claim 9 , the processing unit to further:
in response to a request for the at least one of the plurality of subscriber encoded content files, identify the top-k set of publisher encoded content files.
11. The computing platform as recited in claim 9 , the processing unit to further:
determine the top-k set of publisher encoded content files from at least the publisher encoded content files to be ranked.
12. The computing platform as recited in claim 9 , the processing unit to further:
determine the top-k set of publisher encoded content files to comprise use of a top-k retrieval publish-subscribe process to comprise at least one of: a term-at-a-time (TAAT) publish-subscribe process; a skipping TAAT publish-subscribe process; a document-at-a-time (DAAT) publish-subscribe process; or a skipping DAAT publish-subscribe process.
13. The computing platform of claim of claim 9 , the processing unit to generate the indices to be based at least in part on an indication of content relevancy.
14. The computing platform of claim 9 , wherein to replace one publisher encoded content file in the top-k set of publisher encoded content files is to: compare the ranking score of the new publisher encoded content file with the ranking score of one or more publisher encoded content files of the top-k set of publisher encoded content files, and responsive to the comparison, remove a publisher encoded content file with a lower ranking score from the top-k set of publisher encoded content files and add the new publisher encoded content file to the top-k set of publisher encoded content files.
15. An article computing:
a non-transitory computer readable medium having stored therein computer implementable instructions executable by a processing unit to:
generate and maintain a subscriber index for one or more of a plurality of subscriber encoded content files to comprise at least informational story content, and a publisher index for one or more of the plurality of publisher encoded content files to comprise at least micro-blog and/or social network content;
determine a content-based score and a recency score for the one or more of the plurality of publisher encoded content files for association with the one or more of the plurality of subscriber encoded content files, and use, at least in part, the to be determined content-based score and the to be determined recency score to determine a ranking score for the one or more of the plurality of publisher encoded content files for the association with the one or more of the plurality of subscriber encoded content files;
determine for the one or more of the plurality of subscriber encoded content files a top-k set of publisher encoded content files to be based, at least in part, on the to be determined ranking score;
in response to a new publisher encoded content file:
query the subscriber index, at least a portion of the new publisher encoded content file to be employed to determine a ranking score for the new publisher encoded content file with regard to the one or more of the plurality of subscriber encoded content files, wherein the ranking score is to be based, at least in part, on a content-based score and a recency score, wherein the recency score is to be based at least in part on a difference between a time of page view and a time of creation of the new publisher encoded content file; and
determine whether the new publisher encoded content file is to replace one publisher encoded content file in the top-k set of publisher encoded content files to be based, at least in part, on a comparison of the ranking score of the new publisher encoded content file and the ranking score of the top-k set of publisher encoded content files.
16. The article as recited in claim 15 , the computer implementable instructions being further executable by the processing unit to:
in response to a request for the at least one of the plurality of subscriber encoded content files, identify the top-k set of publisher encoded content files.
17. The article as recited in claim 15 , the computer implementable instructions being further executable by the processing unit to:
determine the top-k set of publisher encoded content files from at least the publisher encoded content files to be ranked.
18. The article as recited in claim 15 , the computer implementable instructions being further executable by the processing unit to:
determine the top-k set of publisher encoded content files to comprise use of a top-k retrieval publish-subscribe process to comprise at least one of: a term-at-a-time (TAAT) publish-subscribe process; a skipping TAAT publish-subscribe process; a document-at-a-time (DAAT) publish-subscribe process; or a skipping DAAT publish-subscribe process.
19. The article of claim 15 , wherein the instructions are further executable to generate the indices to be based at least in part on an indication of content relevancy.
20. The article of claim 15 , wherein to replace one publisher encoded content file in the top-k set of publisher encoded content files the computer implementable instructions are to be further executable by the processing unit to: compare the ranking score of the new publisher encoded content file with the ranking score of one or more publisher encoded content files of the top-k set of publisher encoded content files, and responsive to the comparison, remove a publisher encoded content file with a lower ranking score from the top-k set of publisher encoded content files and add the new publisher encoded content file to the top-k set of publisher encoded content files.