Classifying social entities and applying unique policies on social entities based on crowd-sourced data
Technology is disclosed for detecting, classifying, and/or enforcing rules on social networking activity. The technology can scan and collect social content data from one or more social networks, store the social content data, classify content data posted to a social network, create and apply a set of social data content rules to future posted social content data.
1. A method, comprising:
detecting, by a computer, a comment posted, uploaded, or being posted or uploaded to a first social account of a social network on Internet;
determining, by the computer, whether the comment is recognizable as known undesirable content;
responsive to a determination that the comment is not recognizable as known undesirable content, determining, by the computer, whether the comment is from a known undesirable source;
responsive to the comment not being from a known undesirable source, determining, by the computer, whether the comment contains content similar to undesirable content;
responsive to the comment not being from a known undesirable source and not containing content similar to undesirable content, determining, by the computer, whether a source of the comment is similar to any known undesirable source;
responsive to the source of the comment not being similar to any known undesirable source, and the comment not being from a known undesirable source, not containing content similar to undesirable content, and not recognizable as known undesirable content, flagging or marking, by the computer, the comment as not undesirable content; and
responsive to the source of the comment being similar to any known undesirable source, or the comment being from a known undesirable source, containing content similar to undesirable content, or being recognizable as known undesirable content, flagging or marking, by the computer, the comment as undesirable content.
2. The method according to claim 1 , further comprising:
responsive to the comment being flagged or marked as undesirable content, performing at least one of:
classifying social data subsequently posted by the source of the comment to the first social account as undesirable content;
changing a sensitivity to undesirable content classification on subsequent social data posted by the source of the comment to the first social account;
classifying subsequent social data that is same or similar to the comment posted to the first social account as containing undesirable content regardless of source; or
considering subsequent social data posted to the first social account, by any source that is same or similar to the source of the comment, as containing undesirable content.
3. The method according to claim 1 , wherein the source of the comment comprises the first social account, another social account of the social network, or a third-party application posting on behalf of the first social account.
4. The method according to claim 1 , further comprising:
evaluating, by an inference engine relative to a specified threshold, a frequency or percentage of undesirable content present in the social data posted by the first social account;
based on the evaluating, determining, by the inference engine, that undesirable content is present in the social data posted by the first social account;
responsive to a determination that undesirable content is present in the social data posted by the first social account, informing, by the inference engine, a classification engine to take one or more actions or apply one or more rules to all social data classified by the classification engine.
5. The method according to claim 4 , wherein the classification engine is adapted to perform at least one of:
classifying social data subsequently posted by the source of the comment to the first social account as undesirable content;
changing a sensitivity to undesirable content classification on subsequent social data posted by the source of the comment to the first social account;
classifying subsequent social data that is same or similar to the comment posted to the first social account as containing undesirable content regardless of source; or
considering subsequent social data posted to the first social account, by any source that is same or similar to the source of the comment, as containing undesirable content.
6. The method according to claim 1 , wherein the undesirable content comprises at least one of: spam, linked content containing a universal resource locator, profanity, abusive speech, malware, bullying content, fraudulent activity, potentially violent speech, sexually predatory activity, or malicious content.
7. The method according to claim 1 , further comprising:
performing the detecting at a regular time interval, at a random time interval, upon detection of new content being posted or uploaded or about to be posted or uploaded to the first social account via an application programming interface of the social network on the Internet.
8. A system, comprising:
a processor;
a non-transitory computer-readable medium; and
stored instructions translatable by the processor for:
detecting a comment posted, uploaded, or being posted or uploaded to a first social account of a social network on Internet;
determining whether the comment is recognizable as known undesirable content;
responsive to a determination that the comment is not recognizable as known undesirable content, determining whether the comment is from a known undesirable source;
responsive to the comment not being from a known undesirable source, determining whether the comment contains content similar to undesirable content;
responsive to the comment not being from a known undesirable source and not containing content similar to undesirable content, determining whether a source of the comment is similar to any known undesirable source;
responsive to the source of the comment not being similar to any known undesirable source, and the comment not being from a known undesirable source, not containing content similar to undesirable content, and not recognizable as known undesirable content, flagging or marking the comment as not undesirable content; and
responsive to the source of the comment being similar to any known undesirable source, or the comment being from a known undesirable source, containing content similar to undesirable content, or being recognizable as known undesirable content, flagging or marking the comment as undesirable content.
9. The system of claim 8 , wherein the stored instructions are further translatable by the processor for:
responsive to the comment being flagged or marked as undesirable content, performing at least one of:
classifying social data subsequently posted by the source of the comment to the first social account as undesirable content;
changing a sensitivity to undesirable content classification on subsequent social data posted by the source of the comment to the first social account;
classifying subsequent social data that is same or similar to the comment posted to the first social account as containing undesirable content regardless of source; or
considering subsequent social data posted to the first social account, by any source that is same or similar to the source of the comment, as containing undesirable content.
10. The system of claim 8 , wherein the source of the comment comprises the first social account, another social account of the social network, or a third-party application posting on behalf of the first social account.
11. The system of claim 8 , wherein the stored instructions are further translatable by the processor for:
evaluating, by an inference engine relative to a specified threshold, a frequency or percentage of undesirable content present in the social data posted by the first social account;
based on the evaluating, determining, by the inference engine, that undesirable content is present in the social data posted by the first social account;
responsive to a determination that undesirable content is present in the social data posted by the first social account, informing, by the inference engine, a classification engine to take one or more actions or apply one or more rules to all social data classified by the classification engine.
12. The system of claim 11 , wherein the classification engine is adapted to perform at least one of:
classifying social data subsequently posted by the source of the comment to the first social account as undesirable content;
changing a sensitivity to undesirable content classification on subsequent social data posted by the source of the comment to the first social account;
classifying subsequent social data that is same or similar to the comment posted to the first social account as containing undesirable content regardless of source; or
considering subsequent social data posted to the first social account, by any source that is same or similar to the source of the comment, as containing undesirable content.
13. The system of claim 8 , wherein the undesirable content comprises at least one of: spam, linked content containing a universal resource locator, profanity, abusive speech, malware, bullying content, fraudulent activity, potentially violent speech, sexually predatory activity, or malicious content.
14. The system of claim 8 , wherein the stored instructions are further translatable by the processor for:
performing the detecting at a regular time interval, at a random time interval, upon detection of new content being posted or uploaded or about to be posted or uploaded to the first social account via an application programming interface of the social network on the Internet.
15. A computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor for:
detecting a comment posted, uploaded, or being posted or uploaded to a first social account of a social network on Internet;
determining whether the comment is recognizable as known undesirable content;
responsive to a determination that the comment is not recognizable as known undesirable content, determining whether the comment is from a known undesirable source;
responsive to the comment not being from a known undesirable source, determining whether the comment contains content similar to undesirable content;
responsive to the comment not being from a known undesirable source and not containing content similar to undesirable content, determining whether a source of the comment is similar to any known undesirable source;
responsive to the source of the comment not being similar to any known undesirable source, and the comment not being from a known undesirable source, not containing content similar to undesirable content, and not recognizable as known undesirable content, flagging or marking the comment as not undesirable content; and
responsive to the source of the comment being similar to any known undesirable source, or the comment being from a known undesirable source, containing content similar to undesirable content, or being recognizable as known undesirable content, flagging or marking the comment as undesirable content.
16. The computer program product of claim 15 , wherein the instructions are further translatable by the processor for:
responsive to the comment being flagged or marked as undesirable content, performing at least one of:
classifying social data subsequently posted by the source of the comment to the first social account as undesirable content;
changing a sensitivity to undesirable content classification on subsequent social data posted by the source of the comment to the first social account;
classifying subsequent social data that is same or similar to the comment posted to the first social account as containing undesirable content regardless of source; or
considering subsequent social data posted to the first social account, by any source that is same or similar to the source of the comment, as containing undesirable content.
17. The computer program product of claim 15 , wherein the source of the comment comprises the first social account, another social account of the social network, or a third-party application posting on behalf of the first social account.
18. The computer program product of claim 15 , wherein the instructions are further translatable by the processor for:
evaluating, by an inference engine relative to a specified threshold, a frequency or percentage of undesirable content present in the social data posted by the first social account;
based on the evaluating, determining, by the inference engine, that undesirable content is present in the social data posted by the first social account;
responsive to a determination that undesirable content is present in the social data posted by the first social account, informing, by the inference engine, a classification engine to take one or more actions or apply one or more rules to all social data classified by the classification engine, wherein the classification engine is adapted to perform at least one of:
classifying social data subsequently posted by the source of the comment to the first social account as undesirable content;
changing a sensitivity to undesirable content classification on subsequent social data posted by the source of the comment to the first social account;
classifying subsequent social data that is same or similar to the comment posted to the first social account as containing undesirable content regardless of source; or
considering subsequent social data posted to the first social account, by any source that is same or similar to the source of the comment, as containing undesirable content.
19. The computer program product of claim 15 , wherein the undesirable content comprises at least one of: spam, linked content containing a universal resource locator, profanity, abusive speech, malware, bullying content, fraudulent activity, potentially violent speech, sexually predatory activity, or malicious content.
20. The computer program product of claim 15 , wherein the instructions are further translatable by the processor for:
performing the detecting at a regular time interval, at a random time interval, upon detection of new content being posted or uploaded or about to be posted or uploaded to the first social account via an application programming interface of the social network on the Internet.