INFORMATION IDENTIFICATION AND EXTRACTION

1 . A computer implemented method of using social media accounts as sensors, the method comprising:

obtaining a set of seed social media accounts based on the set of seed social media accounts being associated with known contributors to a topic;

for each seed social media account of the set of seed social media accounts:

fetching a social media account profile;

extracting a first set of social media accounts associated with the seed social media account from the social media account profile, the seed social media account obtaining information from the first set of social media accounts; and

based on a comparison of the first set of social media accounts with the set of seed social media accounts, constructing a first candidate set of social media accounts that includes social media accounts in the first set of social media accounts and not in the seed set of social media accounts;

for each candidate social media account of the first candidate set of social media accounts:

analyzing the candidate social media account to determine a type of social media account, the type of social media account including knowledgeable social media accounts and learning social media accounts; and

in response to the candidate social media account being determined to be a knowledgeable social media account, adding the candidate social media account to the set of seed social media accounts; and

providing content to the set of seed social media accounts, the content selected based on the topic.

2 . The method of claim 1 , wherein analyzing the candidate social media account to determine a type of social media account includes:

obtaining a name associated with the candidate social media account;

determining whether the name includes a person name or an organization name;

in response to determining that the name includes an organization name, removing the candidate account from the first set of candidate social media accounts;

in response to determining that the name includes a person name, comparing the name with an author database of publications related to the topic; and

in response to the name matching a name in the author database, adding the candidate social media account to the seed set of social media accounts.

3 . The method of claim 2 , wherein analyzing the candidate social media account to determine a type of social media account further comprises:

in response to determining that the name does not match a name in the author database, identifying an interest score for the candidate account relative to the topic based on topics from postings on the candidate social media account and third party interactions with the postings; and

in response to determining that the interest score exceeds a threshold, adding the candidate social media account to the seed set of social media accounts.

4 . The method of claim 1 , further comprising, for each seed social media account of the set of seed social media accounts:

extracting a second set of social media accounts associated with the seed social media account from the social media account profile, the second set of social media accounts obtaining information from the seed social media account;

based on a comparison of the second set of social media accounts with the set of seed social media accounts, constructing a second candidate set of social media accounts that includes social media accounts in the second set of social media accounts and not in the seed set of social media accounts or the first set of social media accounts;

for each second candidate social media account of the second candidate set of social media accounts:

obtaining a name associated with the second candidate social media account;

determining whether the name includes a person name or an organization name;

in response to determining that the name is a person name, identifying an interest score for the second candidate account relative to the topic based on topics from postings on the second candidate social media account and interactions with third party postings, or a combination thereof; and

in response to determining that the interest score exceeds a threshold, adding the second candidate social media account to a second set of social media accounts; and

providing second content to the second set of social media accounts, the second content selected based on the topic, the second content different from the content provided to the seed set of social media accounts.

5 . The method of claim 4 , further comprising removing a generic social media account from the second set of candidate social media accounts.

6 . The method of claim 5 , wherein removing a generic social media account from the second set of candidate social media accounts comprises, for each second candidate social media account of the second set of candidate social media accounts:

fetching the second candidate social media account profile;

tallying a first count of a global set of social media accounts, the global set of social media accounts including all social media accounts that obtain information from the second candidate social media account according to the second candidate social media account profile;

tallying a second count of a local set of social media accounts, the local set of social media accounts including social media accounts in the set of seed social media accounts and the first set of social media accounts that obtain information from the second candidate social media account according to the second candidate social media account profile;

generating a topic-relevancy score by dividing the second count by the first count; and

in response to determining that the topic-relevancy score is below a threshold, removing the second candidate social media account from the second set of social media accounts, the second set of candidate social media accounts, or both.

7 . The method of claim 1 , further comprising removing a generic social media account from the first set of candidate social media accounts.

8 . The method of claim 7 , wherein removing a generic social media account from the first set of candidate social media accounts comprises, for each candidate social media account of the first set of candidate social media accounts:

fetching the candidate social media account profile;

tallying a first count of a global set of social media accounts, the global set of social media accounts including all social media accounts that obtain information from the candidate social media account according to the candidate social media account profile;

tallying a second count of a local set of social media accounts, the local set of social media accounts including social media accounts in the set of seed social media accounts and the first set of social media accounts that obtain information from the candidate social media account according to the candidate social media account profile;

generating a topic-relevancy score by dividing the second count by the first count; and

in response to determining that the topic-relevancy score is below a threshold, removing the candidate social media account from the set of seed social media accounts, the first set of candidate social media accounts, or both.

9 . A non-transitory computer-readable medium containing instructions that, when executed by one or more processors, are configured to perform and/or control performance of operations, the operations comprising:

obtaining a set of seed social media accounts based on the set of seed social media accounts being associated with known contributors to a topic;

for each seed social media account of the set of seed social media accounts:

fetching a social media account profile;

extracting a first set of social media accounts associated with the seed social media account from the social media account profile, the first set of social media accounts obtaining information from the seed social media account; and

based on a comparison of the first set of social media accounts with the set of seed social media accounts, constructing a first candidate set of social media accounts that includes social media accounts in the first set of social media accounts and not in the seed set of social media accounts;

for each candidate social media account of the first candidate set of social media accounts:

analyzing the candidate social media account to determine a type of social media account, the type of social media account including knowledgeable social media accounts and learning social media accounts; and

in response to the candidate social media account being determined to be a knowledgeable social media account, adding the candidate social media account to the set of seed social media accounts; and

providing content to the set of seed social media accounts, the content selected based on the topic.

10 . The computer-readable medium of claim 9 , wherein analyzing the candidate social media account to determine a type of social media account includes:

obtaining a name associated with the candidate social media account;

determining whether the name includes a person name or an organization name;

in response to determining that the name includes an organization name, removing the candidate account from the first set of candidate social media accounts;

in response to determining that the name includes a person name, comparing the name with an author database of publications related to the topic; and

in response to the name matching a name in the author database, adding the candidate social media account to the seed set of social media accounts.

11 . The computer-readable medium of claim 9 , wherein analyzing the candidate social media account to determine a type of social media account further comprises:

in response to determining that the name does not match a name in the author database, identifying an interest score for the candidate account relative to the topic based on topics from postings on the candidate social media account and third party interactions with the postings; and

in response to determining that the interest score exceeds a threshold, adding the candidate social media account to the seed set of social media accounts.

12 . The computer-readable medium of claim 9 , the operations further comprising, for each seed social media account of the set of seed social media accounts:

extracting a second set of social media accounts associated with the seed social media account from the social media account profile, the seed social media account obtaining information from the second set of social media accounts;

based on a comparison of the second set of social media accounts with the set of seed social media accounts, constructing a second candidate set of social media accounts that includes social media accounts in the second set of social media accounts and not in the seed set of social media accounts or the first set of social media accounts;

for each second candidate social media account of the second candidate set of social media accounts:

obtaining a name associated with the second candidate social media account;

determining whether the name includes a person name or an organization name;

in response to determining that the name is a person name, identifying an interest score for the second candidate account relative to the topic based on topics from postings on the second candidate social media account and interactions with third party postings, or a combination thereof; and

in response to determining that the interest score exceeds a threshold, adding the second candidate social media account to a second set of social media accounts; and

providing second content to the second set of social media accounts, the second content selected based on the topic, the second content different from the content provided to the seed set of social media accounts.

13 . The computer-readable medium of claim 12 , the operations further comprising further comprising removing a generic social media account from the second set of candidate social media accounts.

14 . The computer-readable medium of claim 13 , wherein removing a generic social media account from the second set of candidate social media accounts comprises, for each second candidate social media account of the second set of candidate social media accounts:

fetching the second candidate social media account profile;

tallying a first count of a global set of social media accounts, the global set of social media accounts including all social media accounts that obtain information from the second candidate social media account according to the second candidate social media account profile;

tallying a second count of a local set of social media accounts, the local set of social media accounts including social media accounts in the set of seed social media accounts and the first set of social media accounts that obtain information from the second candidate social media account according to the second candidate social media account profile;

generating a topic-relevancy score by dividing the second count by the first count; and

in response to determining that the topic-relevancy score is below a threshold, removing the second candidate social media account from the second set of social media accounts, the second set of candidate social media accounts, or both.

15 . The computer-readable medium of claim 9 , the operations further comprising removing a generic social media account from the first set of candidate social media accounts.

16 . The computer-readable medium of claim 15 , wherein removing a generic social media account from the first set of candidate social media accounts comprises, for each candidate social media account of the first set of candidate social media accounts:

fetching the candidate social media account profile;

tallying a first count of a global set of social media accounts, the global set of social media accounts including all social media accounts that obtain information from the candidate social media account according to the candidate social media account profile;

tallying a second count of a local set of social media accounts, the local set of social media accounts including social media accounts in the set of seed social media accounts and the first set of social media accounts that obtain information from the candidate social media account according to the candidate social media account profile;

generating a topic-relevancy score by dividing the second count by the first count; and

in response to determining that the topic-relevancy score is below a threshold, removing the candidate social media account from the set of seed social media accounts, the first set of candidate social media accounts, or both.

17 . A system comprising:

one or more social media servers; and

a computing device including:

one or more processors, and

a non-transitory computer-readable medium containing instructions that, when executed by the one or more processors, are configured to perform and/or control performance of operations, the operations comprising:

obtaining a set of seed social media accounts based on the set of seed social media accounts being associated with known contributors to a topic;

for each seed social media account of the set of seed social media accounts:

fetching a social media account profile from the one or more social media servers;

extracting a first set of social media accounts associated with the seed social media account from the social media account profile, the first set of social media accounts obtaining information from the seed social media account; and

based on a comparison of the first set of social media accounts with the set of seed social media accounts, constructing a first candidate set of social media accounts that includes social media accounts in the first set of social media accounts and not in the seed set of social media accounts;

for each candidate social media account of the first candidate set of social media accounts:

analyzing the candidate social media account to determine a type of social media account, the type of social media account including knowledgeable social media accounts and learning social media accounts; and

in response to the candidate social media account being determined to be a knowledgeable social media account, adding the candidate social media account to the set of seed social media accounts; and

providing content to the set of seed social media accounts, the content selected based on the topic.

18 . The system of claim 17 , the operations further comprising, for each seed social media account of the set of seed social media accounts:

extracting a second set of social media accounts associated with the seed social media account from the social media account profile, the seed social media account obtaining information from the second set of social media accounts;

based on a comparison of the second set of social media accounts with the set of seed social media accounts, constructing a second candidate set of social media accounts that includes social media accounts in the second set of social media accounts and not in the seed set of social media accounts or the first set of social media accounts;

for each second candidate social media account of the second candidate set of social media accounts:

obtaining a name associated with the second candidate social media account;

determining whether the name includes a person name or an organization name;

in response to determining that the name is a person name, identifying an interest score for the second candidate account relative to the topic based on topics from postings on the second candidate social media account and interactions with third party postings, or a combination thereof; and

in response to determining that the interest score exceeds a threshold, adding the second candidate social media account to a second set of social media accounts; and

providing second content to the second set of social media accounts, the second content selected based on the topic, the second content different from the content provided to the seed set of social media accounts.

19 . The system of claim 18 , the operations further comprising, for each second candidate social media account of the second set of candidate social media accounts:

fetching the second candidate social media account profile from the one or more social media servers;

tallying a first count of a global set of social media accounts, the global set of social media accounts including all social media accounts that obtain information from the second candidate social media account according to the second candidate social media account profile;

tallying a second count of a local set of social media accounts, the local set of social media accounts including social media accounts in the set of seed social media accounts and the first set of social media accounts that obtain information from the second candidate social media account according to the second candidate social media account profile;

generating a topic-relevancy score by dividing the second count by the first count; and

in response to determining that the topic-relevancy score is below a threshold, removing the second candidate social media account from the second set of social media accounts, the second set of candidate social media accounts, or both.

20 . The system of claim 17 , the operations further comprising, for each candidate social media account of the first set of candidate social media accounts:

fetching the candidate social media account profile from the one or more social media servers;

tallying a first count of a global set of social media accounts, the global set of social media accounts including all social media accounts that obtain information from the candidate social media account according to the candidate social media account profile;

tallying a second count of a local set of social media accounts, the local set of social media accounts including social media accounts in the set of seed social media accounts and the first set of social media accounts that obtain information from the candidate social media account according to the candidate social media account profile;

generating a topic-relevancy score by dividing the second count by the first count; and

in response to determining that the topic-relevancy score is below a threshold, removing the candidate social media account from the set of seed social media accounts, the first set of candidate social media accounts, or both.