Discovering and ranking trending links about topics

• US 9,009,126 B2
• Filed: 03/15/2013
• Issued: 04/14/2015
• Est. Priority Date: 07/31/2012
• Status: Active Grant

• Current Assignee: Blazent Incorporated (ServiceNow Incorporated)
• Original Assignee: Bottlenose, Inc. (ServiceNow Incorporated)
• Inventors: Spivack, Nova, Heide, Dominiek ter
• Primary Examiner(s): NGUYEN, KIM T
• Application Number:
  US13/844,588
• Publication Number:
  US 20140040281A1
• Time in Patent Office: 760 Days
• Field of Search: 707/705, 707/748
• US Class Current: 707/705
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: H04L 51/52 for supporting social networking services

Abstract

A system and a method for discovering and ranking trending links about topics are presented. The method comprises steps of receiving a plurality of messages from a social networking server, identifying a plurality of trending objects from the plurality of messages, generating at least one trending score for each trending object of the trending objects, and presenting a list of the trending objects based on the trending scores.

Claims

1. A method for increasing a machine'"'"'s accuracy in analyzing trends within a data stream received by the machine, the method comprising:

receiving, at a social intelligence system, a plurality of messages from a social networking server;

identifying, using a processor of the system, at least two trending objects from the plurality of messages;

generating, using the processor, a momentum score for each of the plurality of trending objects, wherein the momentum score is calculated based on a boost factor that;

(i) exponentially decreases from a maximum boost value to a value of one (1) in a predetermined period of time starting from when an associated message is received, and (ii) continues to exponentially decrease until the associated message expires;

generating, using the processor, a co-occurrence score for the two trending objects, the co-occurrence score is calculated based on a number of messages that mention both of the two trending objects;

associating, using the processor, the two trending objects if the co-occurrence score exceeds a predetermined value; and

generating, using the processor, a plurality of trending scores for the trending objects, wherein a trending score of a respective trending object is generated at least based on the respective trending object'"'"'s momentum score, and wherein the generation of the trending score of the respective trending object treats the two associated trending objects as a single trending object.

2. The method of claim 1, further comprising:

treating the two associated trending objects as a single trending object for predicting trends based on the plurality of messages.

3. The method of claim 1, wherein the trending scores are normalized to a predetermined scale.

4. The method of claim 1, wherein the trending scores are generated irrespective of any specific user of the social networking server.

5. The method of claim 1, wherein the trending objects represent a plurality of objects in a category.

6. The method of claim 5, wherein the objects in the category include movies, TV shows, advertisements, online videos, celebrities, news articles, brands, products, or photos.

7. The method of claim 1, wherein the momentum score of the trending object equals M, wherein the maximum boost value equals maxBoost, and wherein M=maxBoost/message age^(Log(maxBoost)/Log(the predetermined period of time)).

8. The method of claim 1, further comprising:

receiving, at the social intelligence system, a query from a user;

generating, using the processor, an importance score for each of the plurality of messages, wherein the importance score is calculated based on a multiplication product of (i) the momentum scores of all the identified trending objects that correspond to a given message, (ii) a total number of available links to the given message, (iii) a first relevancy between content of the given message and the query, (iv) a second relevancy between the trending objects that correspond to the given message and an interest profile of the user, and (v) a number of subscribers who follow an author of the given message; and

ranking the plurality of messages based on their importance scores.

9. The method of claim 1, wherein the trending score of the respective trending object further depends on a number of times that the trending object appears in the plurality of messages.

10. The method of claim 1, wherein the trending score of the respective trending object further depends on a time period after a latest message of the plurality of messages mentioned the trending object is received.

11. The method of claim 1, wherein the trending score of the respective trending object further depends on a number of followers of a particular user or a set of users who mentioned the trending object in a message of the plurality of messages,

The method of claim 1, wherein the trending score of the respective trending object further depends on a number of participants in a particular conversation or who interacted with contents related to a particular topic or set of topics.

12. The method of claim 1, wherein the trending score of the respective trending object further depends on an influence profile of a user who mentioned the trending object in a message of the plurality of messages.

13. The method of claim 1, wherein the trending score of the respective trending object further depends on whether the plurality of messages includes the trending object or information related to the trending object.

14. The method of claim 1, wherein the trending objects include topics, types, hashtags, people, messages, or links.

15. The method of claim 1, further comprising:

presenting the trending objects as scatters or paths in a quadrant scatter plot, wherein a volume of each scatter indicates a trending score of a trending object represented by the scatter.

16. The method of claim 1, wherein said identifying includes:

identifying a plurality of trending objects from the plurality of messages using a speech tagging process.