METHOD AND APPARATUS FOR CONTENT-VIRALITY AMPLIFICATION
A method and apparatus for inducing expansion of audience of source contents posted in a social medium are disclosed. Upon posting the source contents, spontaneous audience are detected and content-coupled audience are determined; the content-coupled audience and the spontaneous audience are users of the social medium that accessed common contents, or contents of a common class, other than the source contents. A set of content-coupled audience is pruned to determine a set of prospective audience. Information relevant to the source contents is disseminated to the prospective audience resulting in audience gain. Processes of iterative-recursive exploration of multi-stratum content coupling are performed to further enhance the audience gain. An apparatus implementing the method may employ a single processor or multiple processors operating concurrently and exchanging intermediate results through commonly accessed buffers. Several variations of the method applying different criteria for pruning commonly-accessed content as well as content-coupled audience are disclosed.
1 . A method of audience amplification comprising:
employing at least one processor for:
posting source contents in a social medium serving a plurality of users and providing a plurality of contents;
detecting a set of spontaneous audience that individually accessed at least one of said source contents;
determining first-stratum contents that said spontaneous audience accessed;
content-pruning said first-stratum contents to yield first-stratum relevant contents;
detecting first-stratum users that accessed said first-stratum relevant contents;
user-pruning said first-stratum users to yield first-stratum prospective audience;
communicating said source contents to said first-stratum prospective audience; and
tracking said first-stratum prospective audience to identify first-stratum induced audience.
2 . The method of claim 1 wherein said determining comprises forming said first-stratum contents as a union of sets of contents accessed by individual users of said spontaneous audience.
3 . The method of claim 1 wherein said detecting comprises forming said first-stratum users as a union of sets of users that accessed individual contents of said first-stratum relevant contents.
4 . The method of claim 1 further comprising:
determining a gravitation score of each of said first-stratum contents; and
eliminating each first-stratum content having a gravitation score below a predefined gravitation threshold to yield said first-stratum relevant contents;
said gravitation score being a count of users that accessed said each first-stratum content.
5 . The method of claim 1 wherein said content-pruning comprises:
determining a content-similarity level of each of said first-stratum contents to said source contents; and
eliminating each first-stratum content having a content-similarity level below a predefined content-similarity threshold.
6 . The method of claim 1 wherein said user-pruning further comprises:
determining an activity score of each of said first-stratum users; and
eliminating each first-stratum user having an activity score below a predefined activity threshold;
said activity score being a count of contents of said first-stratum relevant contents that said each first-stratum user accessed.
7 . The method of claim 6 wherein said user-pruning further comprises:
determining a user-similarity level of each of said first-stratum users to said spontaneous audience; and
eliminating each first-stratum user having a user-similarity level below a predefined user-similarity threshold.
8 . The method of claim 6 further comprising:
segmenting said first-stratum users into a predefined number of clusters of users; and
selecting one of the clusters of users as a relevant cluster.
9 . The method of claim 8 wherein said user-pruning further comprises retaining only first-stratum users within said relevant cluster to yield said first-stratum prospective audience.
10 . The method of claim 8 wherein said segmenting is based on characterizing each of said first-stratum users by respective descriptors of predefined descriptor types.
11 . The method of claim 6 wherein said user-pruning further comprises:
determining a user-similarity level of each user of said first-stratum users to a predefined set of model consumers; and
eliminating each first-stratum user having a user-similarity level below a predefined user-similarity threshold.
12 . (canceled)
13 . The method of claim 1 wherein said communicating comprises multicasting said source contents to said first-stratum prospective audience.
14 . The method of claim 1 further comprising:
tracking said first-stratum prospective audience;
determining second-stratum contents that said first-stratum prospective audience accessed;
content-pruning said second-stratum contents to yield second-stratum relevant contents;
detecting second-stratum users that accessed said second-stratum relevant contents;
user-pruning said second-stratum users to yield second-stratum prospective audience;
communicating said source contents to said second-stratum prospective audience; and
tracking said second-stratum prospective audience to identify second-stratum induced audience.
15 . The method of claim 1 further comprising:
tracking said first-stratum induced audience;
determining second-stratum contents that said first-stratum induced audience accessed;
content-pruning said second-stratum contents to yield second-stratum relevant contents;
detecting second-stratum users that accessed said second-stratum relevant contents;
user-pruning said second-stratum users to yield second-stratum prospective audience;
communicating said source contents to said second-stratum prospective audience; and
tracking said second-stratum prospective audience to identify second-stratum induced audience.
16 . A method of audience amplification comprising:
employing at least one processor for:
posting source contents in a social medium serving a plurality of users and providing a plurality of contents;
detecting, over a specified time interval, a set of spontaneous audience that individually accessed at least one of said source contents;
populating a set of augmented audience to comprise said set of spontaneous audience; and
performing processes of:
determining first-stratum contents that said augmented audience accessed;
content-pruning said first-stratum contents to yield relevant contents;
detecting first-stratum users that accessed said relevant contents;
user-pruning said first-stratum users to yield prospective audience;
communicating said source contents to said prospective audience;
tracking said prospective audience to identify induced audience; and
adding said induced audience to said set of augmented audience.
17 . The method of claim 16 further comprising:
setting a cyclic timer to restart after a predefined interval of time defining a timer cycle;
determining audience gain as a count of said induced audience during each timer cycle;
determining a trend of audience gain over successive timer cycles; and
repeating said performing subject to a determination that said trend satisfies a respective predetermined criterion.
18 . The method of claim 16 wherein said content-pruning comprises processes of:
determining for each of said first-stratum contents;
a gravitation score; and
a level of content-similarity to said source contents;
and
eliminating each first-stratum content having at least one of:
a gravitation score below a predefined gravitation threshold; and
a level of content-similarity below a predefined content-similarity threshold;
said gravitation score being a count of users that accessed said each first-stratum content.
19 . The method of claim 16 wherein said user-pruning comprises:
determining for each of said first-stratum users:
an activity score; and
a level of user-similarity to a predefined set of model consumers;
and
eliminating each first-stratum user having at least one of:
an activity score below a predefined activity threshold; and
a user-similarity level below a predefined user-similarity threshold;
said activity score being a count of said first-stratum relevant contents that said each first-stratum user accessed.
20 - 24 . (canceled)
25 . An apparatus for audience amplification coupled to a social medium serving a plurality of users and providing a plurality of contents, the apparatus comprising:
processors and memory devices storing processor executable instructions causing said processors to:
post a source content in said social medium;
identify initial users that spontaneously accessed said source content over a specified time interval;
populate a set of augmented audience to comprise said set of initial users; and
perform an iterative procedure to:
acquire first-stratum contents that said augmented audience accessed;
filter said first-stratum contents to retain relevant contents;
detect first-stratum users that accessed said relevant contents;
prune said first-stratum users to yield prospective audience;
communicate said source contents to said prospective audience;
track said prospective audience to identify induced audience; and
add said induced audience to said set of augmented audience.
26 . The apparatus of claim 25 further comprising a cyclic timer configured to restart after a predefined interval of time defining a timer cycle, wherein said processor executable instructions cause said processors to:
determine audience gain during each timer cycle;
determine a trend of audience gain over successive timer cycles; and
continue said iterative procedure while said trend satisfies a respective predetermined criterion.