System and method for customizing analytics based on users media affiliation status
A new approach is proposed that contemplates systems and methods to generate customized search results as well as metrics, such as aggregated sentiment, counts of targets or sources or citations, or aggregated gross impressions or exposure, of social media content items over a social network while discriminating between the perspectives of individuals from the media and individuals not from the media. This approach can be used to generate search results and/or metrics including only media perspectives, or excluding media perspectives. More specifically, while social media content items are retrieved from corpus based on certain search criteria, for the purpose of providing search results or providing aggregated metrics, the search criteria can include a media or non-media filter, which is applied to the authors posting social media content to exclude or include certain authors meeting media/non-media criteria. For the application of a non-media filter, content from media authors can be excluded, or ranked below content from non-media authors. Similarly, for a media filter, content from non-media authors can be excluded, or ranked below content from media authors.
1. A system, comprising:
a citation search and analytics engine that includes a processor, which in operation, retrieves from a social network a plurality of content items composed by a plurality of subjects citing a plurality of objects that fit searching criteria specified by a user; and
an object selection engine that includes a processor, which in operation,
receives a user-provided query with selection criteria including a separate media-affiliation status preference, identifies media affiliation status of the plurality of subjects of the plurality of content items meeting the query selection criteria,
uses a whitelist and a trained probabilistic media type classifier either separately or in combination to assign a subject to each of the content items either as a media type or non-media type author, wherein the object selection derives the whitelist from a public list of social media sources and their respective verified accounts, which classify the users/owners of those accounts as either media type or non-media type authors engine,
utilizes the media affiliation status of the plurality of subjects as a preference to select at least one of the plurality of content items or the cited objects such that those items/objects for which the plurality of subjects media affiliation status matches the user-provided media affiliation status preference, wherein the media affiliation status for one of the plurality of subjects is whether that subject is associated with a commercial news source,
provides this selection in an un-ranked search result,
computes and provides aggregated metrics based on this selection, and
ranks the selected content items or cited object higher than other content items/cited objects in a ranked search result, based on whether the corresponding subject for the selected content items or cited object is associated with media affiliation status preference, wherein one of the selected content items or cited objects that is associated with the corresponding subject matching the media affiliation status preference is ranked higher than another one of the content items or cited objects associated with the corresponding subject not matching the media affiliation status preference.
2. The system of claim 1 , wherein:
the social network is a publicly accessible web-based platform or community that enables its users/members to post, share, communicate, and interact with each other.
3. The system of claim 1 , wherein:
the social network is one of, blogs, forums, or any other web-based communities.
4. The system of claim 1 , wherein:
each of the plurality of content items includes one or more of: tweets, replies to the tweets, re-tweets to the tweets, posts, comments to other users' posts, opinions, feeds, connections, references, links to other websites or applications, or any other activities on the social network.
5. The system of claim 1 , wherein:
each of the plurality of subjects is one of:
representation of a person, web log, and entities representing Internet authors or users of social media services, user of microblogging services, users of social networks, reviewer who provides expressions of opinion, reviews, or other information useful for the estimation of influence.
6. The system of claim 1 , wherein:
each of the plurality of objects is one of:
Internet web sites, blogs, videos, books, films, music, image, video, documents, data files, objects for sale, objects that are reviewed or recommended or cited, subjects/authors, natural or legal persons, citations, or any entities that are associated with a Uniform Resource Identifier (URI).
7. The system of claim 1 , wherein:
the citation search and analytics engine sets and adjusts searching, retrieving and ranking criteria and mechanisms of the content items based on user specification and/or internal statistical data.
8. The system of claim 1 , wherein:
the media type of an author of a content item originated from a professional reporting or news agency is classified as a media type author.
9. The system of claim 1 , wherein:
the media type of an individual of a content item not associated with a professional reporting or news agency is classified as a non-media type author.
10. The system of claim 1 , wherein:
the object selection engine trains the media type classifier using an archive of historical content items with clear media type identifications as a training set; and
utilizes the trained the media type classifier to predict the media types of each of the content items with high accuracy.
11. The system of claim 1 , wherein:
the object selection engine reviews profile of the user/author of a content item as well as historical post information by the same user to intelligently identify the media type the author belongs to.
12. The system of claim 1 , wherein:
object selection engine removes content items from one of the plurality subject who have been identified as a media-type author from the search result or ranks such content items lower in the search result.
13. A method, comprising:
retrieving from a social network a plurality of content items composed by a plurality of subjects citing a plurality of objects that fit searching criteria specified by a user;
identifying media affiliation statuses of the plurality of subjects of the plurality of content items;
using a whitelist and a trained probabilistic media type classifier either separately or in combination to assign a subject to each of the content items either as a media type or non-media type author, wherein the whitelist is derived from a public list of social media sources and their respective verified accounts, which classify the users/owners of those accounts as either media type or non-media type authors engine;
utilizing the media affiliation status of the plurality of subjects as a preference to select at least one of the plurality of content items or the cited objects such that those items/objects for which the plurality of subjects media affiliation status matches a user-provided media-affiliation status preference, wherein the media affiliation status for one of the plurality of subjects is whether that subject is associated with a commercial news source;
providing this selection in an un-ranked search result;
computing and providing aggregated metrics based on this selection; and
ranking the selected content items or cited object higher than other content items/cited objects in a ranked search result, based on whether the corresponding subject for the selected content items or cited object is associated with the media affiliation status preference, wherein one of the selected content items or cited objects that is associated with the corresponding subject matching the media affiliation status preference is ranked higher than another one of the content items or cited objects associated with the corresponding subject not matching the media affiliation status preference.
14. The method of claim 13 , further comprising:
setting and adjusting searching, retrieving and ranking criteria and mechanisms of the content items based on user specification and/or internal statistical data.
15. The method of claim 13 , further comprising:
training the media type classifier using an archive of historical content items with clear media type identifications as a training set; and
utilizes the trained the media type classifier to predict the media types of each of the content items with high accuracy.
16. The method of claim 13 , further comprising:
reviewing profile of one of the plurality of subjects of a content item as well as historical post information by the same user to intelligently identify the media type the author belongs to.
17. The method of claim 13 , further comprising:
removing content items from one of the plurality of subjects who have been identified as a media-type author from the search result or ranking such content items lower in the search result.
18. A non-transitory machine-readable medium having executable instructions to cause one or more processing units to perform a method, the method comprising:
retrieving from a social network a plurality of content items composed by a plurality of subjects citing a plurality of objects that fit searching criteria specified by a user;
identifying media affiliation statuses of the plurality of subjects of the plurality of content items;
using a whitelist and a trained probabilistic media type classifier either separately or in combination to assign a subject to each of the content items either as a media type or non-media type author, wherein the whitelist is derived from a public list of social media sources and their respective verified accounts, which classify the users/owners of those accounts as either media type or non-media type authors engine;
utilizing the media affiliation status of the plurality of subjects as a preference to select at least one of the plurality of content items or the cited objects such that only those items/objects for which the plurality of subjects media affiliation status matches a user-provided media-affiliation status preference, wherein the media affiliation status for one of the plurality of subjects is whether that subject is associated with a commercial news source;
providing this selection in an un-ranked search result;
computing and providing aggregated metrics based on this selection; and
ranking the selected content items or cited object higher than other content items/cited objects in a ranked search result, based on whether the corresponding subject for the selected content items or cited object is associated with media affiliation status preference, wherein one of the selected content items or cited objects that is associated with the corresponding subject matching the media affiliation status preference is ranked higher than another one of the content items or cited objects associated with the corresponding subject not matching the media affiliation status preference.