Systems and methods for order-of-magnitude viral cascade prediction in social networks

Patent metadata

Field	Value
Publication number	US-10437945-B2
Application number	US-201615229776-A
Country	US
Kind code	B2
Filing date	Aug 5, 2016
Priority date	Aug 5, 2015
Publication date	Oct 8, 2019
Grant date	Oct 8, 2019

Abstract

Systems and methods for predicting order-of-magnitude viral cascades in social networks are disclosed.

First claim

What is claimed is:

1. A method for processing information associated with social networks, the method comprising:
utilizing a processor, configured for:
generating a social network model comprising a plurality of nodes representing a plurality of users and user devices associated with a social network system and a plurality of edges representing connections between the plurality of users of the social network system;
tracking an adoption through the social network model by the plurality of users of the social network system of a message posted to the social network system for a particular time frame;
applying a community detection algorithm to information associated with the social network model to assign each node of the plurality of nodes to one or multiple communities; and
implementing a system for cascade prediction, including:
providing a plurality of structural diversity measurements, and
computing a classification to predict an order-of-magnitude viral cascade of the message in the social network system based on the tracking of the adoption through the social network model by the plurality of users of the social network system of a message posted to the social network system by:
leveraging the plurality of structural diversity measurements as features, including based on the one or more communities assigned, computing at least some of the features to numerically describe how the plurality of nodes that participate in the cascade are distributed over communities, the features including a measurement of a Gini Impurity and a measure of overlap represented as a number of shared communities associated with the plurality of nodes.

2. The method of claim 1 wherein the adoption of the posted message comprises a reposting of the message to the social network system by a subset of the plurality of users of the social network system.

3. The method of claim 2 further comprising:
comparing the reposting of the message by the subset of the plurality of users of the social network system to a threshold value of reposts of the message.

4. The method of claim 3 wherein the threshold value is 500 reposts of the message.

5. The method of claim 3 wherein the message is predicted as an order-of-magnitude viral message if the tracked reposting of the message by the subset of the plurality of users of the social network system exceeds the threshold value within the particular time frame.

6. The method of claim 5 wherein the particular time frame for tracking the adoption by the subset of the plurality of users of the social network system of the message posted to the social network system through the social network model is 60 minutes.

7. The method of claim 1 wherein the social network system comprises at least one server storing a plurality of webpages accessible by the plurality of users of the social network system.

8. A system for processing information associated with social networks, the method comprising:
a processing device for processing data in a social network system; and
a computer-readable medium associated with the processor and including instructions stored thereon and executable by the processor to:
generate a social network model comprising a plurality of nodes representing a plurality of users of the social network system and a plurality of edges representing connections between the plurality of users of the social network system;
track an adoption through the social network model by the plurality of users of the social network system of a message posted to the social network system for a particular time frame; and
implement a system for cascade prediction that leverages a plurality of structural diversity measurements as features for classification to predict
an order-of-magnitude viral cascade of the message in the social network system based on the tracking of the adoption through the social network model by the plurality of users of the social network system of a message posted to the social network system

wherein the plurality of structural diversity measurements include a number of communities associated with the plurality of nodes, a measure of overlap represented as a number of shared communities associated with the plurality of nodes, a probability of a node being placed in an incorrect community using a measurement of a Gini Impurity, an average time to the adoption by adopters associated with the plurality of nodes, and a number of the plurality nodes.

9. The system of claim 8 wherein the social network system comprises at least one server storing a plurality of webpages accessible by the plurality of users of the social network system.

10. The system of claim 8 wherein the adoption of the posted message comprises a reposting of the message to the social network system by a subset of the plurality of users of the social network system.

11. The system of claim 10 wherein the instructions stored on the computer-readable medium further cause the processor to:
compare the reposting of the message by the subset of the plurality of users of the social network system to a threshold value.

12. The system of claim 11 wherein the threshold value is 500.

13. The system of claim 11 wherein the message is predicted as an order-of-magnitude viral message if the tracked reposting of the message by the subset of the plurality of users of the social network system exceeds the threshold value within the particular time frame.

14. The system of claim 13 wherein the particular time frame for tracking the adoption by the subset of the plurality of users of the social network system of the message posted to the social network system through the social network model is 60 minutes.

15. The method of claim 1 , wherein the plurality of structural diversity measurements include a number of communities associated with the plurality of nodes, a number of shared communities associated with the plurality of nodes, a probability of a node being placed in an incorrect community, an average time to the adoption by adopters associated with the plurality of nodes, and a number of the plurality nodes.

16. The method of claim 1 , wherein the cascade prediction is based solely upon information about a topology associated with the social network model as interpreted by the plurality of structural diversity measurements computed by the system as implemented without consideration of message content.

17. The method of claim 1 , wherein the features define time-based features and size based features.

18. The method of claim 1 , further comprising assigning weights to a subset of the features associated with grouped ones of the plurality of structural diversity measurements.

Assignees

• Univ Arizona State

Inventors

• Shakarian Paulo
• Guo Ruocheng
• Shaabani Elham
• Bhatnagar Abhinav

Classifications

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06F2111/10

  Numerical modelling · CPC title

• G06F30/20 Primary

  Design optimisation, verification or simulation (optimisation, verification or simulation of circuit designs G06F30/30) · CPC title

• G06Q10/04 Primary

  Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem" (market predictions or forecasting for commercial activities G06Q30/0202) · CPC title

• G06Q10/10

  Office automation; Time management · CPC title