Event mining in social networks

• US 8,914,371 B2
• Filed: 12/13/2011
• Issued: 12/16/2014
• Est. Priority Date: 12/13/2011
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Aggarwal, Charu, Subbian, Karthik
• Primary Examiner(s): Uddin, Mohammed R
• Application Number:
  US13/324,513
• Publication Number:
  US 20130151522A1
• Time in Patent Office: 1,099 Days
• Field of Search: 707/737, 707/754, 707/999.107
• US Class Current: 707/737
• CPC Class Codes: G06F 16/24568 Data stream processing; Continuous queries
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

A method and system for detecting an event from a social stream. The method includes the steps of: receiving a social stream from a social network, where the social stream includes at least one object and the object includes a text, sender information of the text, and recipient information of the text; assigning said object to a cluster based on a similarity value between the object and the clusters; monitoring changes in at least one of the clusters; and triggering an alarm when the changes in at least one of the clusters exceed a first threshold value, where at least one of the steps is carried out using a computer device.

Claims

1. A method for detecting an event from a social stream, the method comprising the steps of:

receiving a social stream from a social networks;

wherein said social stream comprises at least one object; and

wherein said object comprises a text, sender information of said text, and recipient information of said text;

clustering each object;

monitoring changes in at least one of said clusters; and

triggering an alarm when said changes in at least one of said clusters exceed a first threshold value;

assigning said object to an existing cluster if a similarity value between said object and said existing cluster is greater than a second threshold value;

creating a new cluster with said object if said similarity value between said object and said existing cluster is less than a second threshold value; and

replacing a stale cluster with said new cluster;

wherein said similarity value is determined by calculating a structural similarity value and a value selected from a group consisting of content-based similarity value, temporal similarity value and combinations thereof,

wherein said structural similarity value is calculated by using structural components of said social network; and

wherein said structural components comprises nodes and node frequencies;

wherein at least one of the steps is carried out using a computer device.

2. The method according to claim 1, wherein said second threshold value is calculated from mean and standard deviation of said similarity value.

3. The method according to claim 1, further comprising the step of:

maintaining said changes in said clusters as historical data of said social stream.

4. The method according to claim 3, wherein said historical data is used for supervised event detection.

5. The method according to claim 4, wherein said step of triggering an alarm further comprises the step of:

utilizing an event signature and a horizon signature.

6. The method according to claim 3, wherein said step maintaining uses a sketch-based technique.

7. The method according to claim 6, wherein said sketch-based technique is used to estimate a structural similarity value between said object and said clusters.

15. A non-transitory computer readable storage medium tangibly embodying a computer readable program code having computer readable instructions which when implemented, cause a computer to carry out the steps of claim 1.

8. A system for detecting an event from a social stream, the system comprising:

a memory;

a processor communicatively coupled to the memory, and a receiving module for receiving a social stream from a social network, communicatively coupled to the memory and the processor, wherein the receiving module is configured to perform the steps of a method comprising;

wherein said social stream comprises at least one object; and

wherein said object comprises a text, sender information of said text, and recipient information of said text;

clustering each object;

a monitoring module for monitoring changes in at least one of said clusters; and

a trigger module for triggering an alarm when said changes in at least one of said clusters exceed a first threshold value;

an existing cluster module for assigning said object to an existing cluster if a similarity value between said object and said existing cluster is greater than a second threshold value;

a new cluster module for creating a new cluster with said object if said similarity value between said object and said existing cluster is less than a second threshold value; and

a replacement module for replacing a stale cluster with said new cluster

wherein said similarity value is determined by calculating a structural similarity value and a value selected from a group consisting of content-based similarity value, temporal similarity value and combinations thereof,

wherein said structural similarity value is calculated by using structural components said social network; and

wherein said structural components comprises nodes and node frequencies.

9. The system according to claim 8, wherein said second threshold value is calculated from mean and standard deviation of said similarity value.

10. The system according to claim 8, further comprising:

a maintenance module for maintaining said changes in said clusters as historical data of said social stream.

11. The system according to claim 10, wherein said historical data is used for supervised event detection.

12. The system according to claim 11, wherein said trigger module further comprises:

a signature module for utilizing an event signature and a horizon signature.

13. The system according to claim 10, wherein said maintenance module uses a sketch-based technique.

14. The system according to claim 13, wherein said sketch-based technique is used to estimate a structural similarity value between said object and said clusters.