Method and system for detection of motivation

1 . A method for media manipulation detection or media threat detection comprising performing a machine-executed operation involving instructions that, when executed by one or more processors, cause the one or more processors to:

direct anonymous users onto the system;

assemble said users into panels, wherein panels are seated for a preset period of time and wherein each preset period of time is divided into a series of rounds, wherein each round occurs at a fixed period of time defining a designated data collection window;

assign each panel a question set wherein at least on question in said question set remains fixed over said preset period of time, wherein said at least one question has the format of a forward looking forecasts, asking about the conduct of others and with a moving time horizon;

in response to a login from a user, send a user interface data, wherein said interface data includes at least one question set with two response fields for each question in said question set, comprising a numerical forecast field and free text entry field;

during said designated data collection window, receive a user response to said question set in the form of an outcome forecast and comments in said free text entry fields, if provided by the user;

store a user tracking number;

store the user response;

in a first round:

compute whether a minimum user participation is achieved;

compute a consensus for each question of said question set based on forecast responses from users during said designated collection window;

compute user forecasts for said designated collection window that fall within preset numerical ranges from said computed consensus;

identify a user comment that corresponds to said computed forecast;

store a direction of influence of said user's forecast relative to the consensus when made;

in response to determining that a forecast-consensus delta exceeds a predetermined threshold, extract a catalyst issue from said identified user comment using linguistic analysis including thesaurus-based concept capture and related term expansion, wherein said catalyst issue is stored in association with said direction of influence of said user's forecast;

store said questions set information, catalyst issue and direction of influence; and

populating a delay list, a positive influence list, or a negative influence list based on said stored catalyst issue and said associated direction of influence;

in rounds subsequent to said first round:

in response to said login from said user with said login credentials, send a user interface data that also includes said user's forecast from said previous round;

compute whether a minimum user participation is achieved;

compute a consensus for each question of said question set based on forecast responses from users during said designated collection window;

compute a relative rate of change between the current and previous round;

compute user forecasts for said designated collection window that fall within preset numerical ranges from said computed consensus;

identify a user comment that corresponds to said computed forecast;

store a direction of influence of said user's forecast relative to the consensus when made;

in response to determining that a forecast-consensus delta exceeds a predetermined threshold, extract a catalyst issue from said identified user comment using linguistic analysis including thesaurus-based concept capture and related term expansion, wherein said catalyst issue is stored in association with said direction of influence of said user's forecast;

store said questions set information, catalyst issue, and direction of influence; and

populating a delay list, a positive influence list, or a negative influence list based on said stored catalyst issue and said associated direction of influence;

store any one of said delay list, said positive influence list, and said negative influence list based on said stored questions set information, catalyst issue, and direction of influence of said rounds subsequent to said first round;

receive a social media post that would otherwise appear in social media but has not yet appeared;

extract terms that appear in said social media post using linguistic analysis including thesaurus-based concept capture and related term expansion; and

perform social media manipulation detection or social media threat detection by determining whether any of the extracted terms appear on the delay list, the positive influence list, and the negative influence list; and

perform social media manipulation countering or social media threat countering by:

buffering the social media post for delayed broadcasting based on a result of the determining of whether any of the extracted terms appear on said delay list;

controlling amplification amplifying the social media post based on the determining of whether any of the extracted terms appear on said positive influence list; and

controlling de-amplification of the social media post based on the determining of whether any of the extracted terms appear on said negative influence list.

2 . The method as claimed in claim 1 , wherein the instructions further cause the one or more processors to: in a subsequent round among the subsequent rounds, assign approximately half of the panel to a control group and assigning a remainder of the panel to a test group, wherein, for said test group said interface data additionally comprises other user comment data or news data; and compute an implicit delta of the test group, wherein the implicit delta is a difference between the question consensus in the current round and the question consensus in the previous round in the test group compared to the difference between the question consensus in the current round and the question consensus in the previous round in the control group.

3 . The method as claimed in claim 1 further comprising the steps of: generating one or more random numbers associated with a question; in response to a request for a random number sending at least one random number to a user; wherein said user's response comprises a user tracking number associated with at least one requested random number; and wherein said user response is stored under said tracking number.

4 . The method as claimed in claim 1 , wherein, in said rounds subsequent to said first round, the computing of the consensus for each question of said question set comprises removing user forecasts of users from said subsequent round consensus who did not have a forecast recorded in the previous round consensus.

5 . The method as claimed in claim 1 , wherein the controlling of the release of the social media post comprises: in response to the result of the determining of whether any of the extracted terms appear on said delay list being that at least one of the extracted terms appears on said delay list, releasing the social media post for consumption if the appearance of the at least one of the extracted terms on said delay list is a first occurrence of the least one of the extracted terms, and marking the social media post for delay if the appearance of the at least one of the extracted terms on said delay list is not the first occurrence of the at least one of the extracted terms.

6 . The method as claimed in claim 5 , wherein a delay associated with social media being marked for delay is computed based on a historic usage frequency of said at least one of the extracted terms, whereby the social media post is not delayed if a number of uses of said at least one of the extracted terms is below the historic usage frequency of said at least one of the extracted terms.

7 . The method as claimed in claim 1 , wherein the controlling of the amplifying of the social media post comprises: in response to the result of the determining of whether any of the extracted terms appear on said positive influence list being that at least one of the extracted terms appears on said positive influence list, amplifying said social media post.

8 . The method as claimed in claim 1 , the controlling of the de-amplifying of the social media post comprises: in response to the result of the determining of whether any of the extracted terms appear on said negative influence list being that at least one of the extracted terms appears on said negative influence list, de-amplifying said social media post.

9 . A data storage system for media manipulation detection or media threat detection, comprising:

one or more processors;

a data store;

wherein said data storage system comprises a user login module that, in response to a login, retrieves for the user a question set and sends a user interface data, wherein said interface data includes at least one question set;

wherein said data storage system further comprises a question module that:

receives a user response to said question in the form of an outcome forecast and comments;

in a first round:

stores whether a minimum user participation is achieved;

stores a consensus for each question of said question set based on forecast responses from users during said designated collection window;

stores user forecasts for said designated collection window that fall within preset numerical ranges from said computed consensus;

stores a user comment that corresponds to said computed forecast;

stores a direction of influence of said user's forecast relative to the consensus when made;

in response to determining that a forecast-consensus delta exceeds a predetermined threshold, extracts a catalyst issue from said identified user comment extracted using linguistic analysis including thesaurus-based concept capture and related term expansion, wherein said catalyst issue is stored in association with said direction of influence of said user's forecast;

stores said questions set information, catalyst issue, and direction of influence; and

populates a delay list, a positive influence list, or a negative influence list based on said stored catalyst issue and said associated direction of influence;

in rounds subsequent to said first round:

stores whether a minimum user participation is achieved;

stores a consensus for each question of said question set based on forecast responses from users during said designated collection window;

stores a relative rate of change between the current and previous round;

stores user forecasts for said designated collection window that fall within preset numerical ranges from said computed consensus;

stores a user comment that corresponds to said computed forecast;

stores a direction of influence of said user's forecast relative to the consensus when made;

in response to determining that a forecast-consensus delta exceeds a predetermined threshold, extracts a catalyst issue from said identified user comment using linguistic analysis including thesaurus-based concept capture and related term expansion, wherein said catalyst issue is stored in association with said direction of influence of said user's forecast;

stores said questions set information, catalyst issue, and direction of influence; and

populates a delay list, a positive influence list, or a negative influence list based on said stored catalyst issue and said associated direction of influence;

wherein said data storage system:

stores any one of a delay list, a positive influence list, and a negative influence list based on said stored questions set information, catalyst issue, and direction of influence of said rounds subsequent to said first round;

receives a social media post that would otherwise appear in social media but has not yet appeared;

extracts terms that appear in said social media post using linguistic analysis including thesaurus-based concept capture and related term expansion;

performs social media manipulation detection or social media threat detection by determining whether any of the extracted terms appear on the delay list, the positive influence list, and the negative influence list; and

performs social media manipulation countering or social media threat countering by:

buffering the social media post for delayed broadcasting based on a result of the determining of whether any of the extracted terms appear on said delay list;

controlling amplification of the social media post based on the determining of whether any of the extracted terms appear on said positive influence list; and

controlling de-amplification of the social media post based on the determining of whether any of the extracted terms appear on said negative influence list.

10 . The data storage system as claimed in claim 9 , wherein: said interface data additionally comprises consensus data, other user comment data, or news data; in a subsequent round, said question module stores approximately half of the panel in a control group and stores a remainder of the panel is stored to a test group, wherein, for said test group said interface data additionally comprises other user comment data or news data; and said question module stores an implicit delta of the test group, where in the implicit delta is a difference between the question consensus in the current round and the question consensus in the previous round in the test group compared to the difference between the question consensus in the current round and the question consensus in the previous round in the control group.

11 . The data storage system as claimed in claim 9 , further comprising a catalysts testing module that: extracts said catalyst issue from said identified user comment; extracts geographical information related to said question; requests use of said catalyst issue in an on-line source associated with said geographic information; identifies at least one isolated spike in use of catalyst information in said on-line source; calculates a consensus for said question and monitoring monitors said consensus for a set number of days following said at least one isolated spike; and determines if said calculated consensus moves in said direction of influence and if so, records the catalyst as a positive for influence.

12 . The data storage system as claimed in claim 9 , wherein the controlling of the release of the social media post comprises: in response to the result of the determining of whether any of the extracted terms appear on said delay list being that at least one of the extracted terms appears on said delay list, releasing the social media post for consumption if the appearance of the at least one of the extracted terms on said delay list is a first occurrence of the least one of the extracted terms, and marking the social media post for delay if the appearance of the at least one of the extracted terms on said delay list is not the first occurrence of the at least one of the extracted terms.

13 . The data storage system as claimed in claim 12 , wherein a delay associated with social media being marked for delay is computed based on a historic usage frequency of said at least one of the extracted terms, whereby the social media post is not delayed if a number of uses of said at least one of the extracted terms is below the historic usage frequency of said at least one of the extracted terms.

14 . The data storage system as claimed in claim 9 , wherein the controlling of the amplifying of the social media post comprises: in response to the result of the determining of whether any of the extracted terms appear on said positive influence list being that at least one of the extracted terms appears on said positive influence list, amplifying said social media post.

15 . The data storage system as claimed in claim 9 , wherein the controlling of the de-amplifying of the social media post comprises: in response to the result of the determining of whether any of the extracted terms appear on said negative influence list being that at least one of the extracted terms appears on said negative influence list, de-amplifying said social media post.