# US 2010/0076913 A1 · Finding communities and their evolutions in dynamic social network

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2010/0076913 A1
- Document ID: US20100076913A1
- Status: Published application
- Kind: A1
- Application number: 12/277,305
- Filed: 2008-11-25
- Published: 2010-03-25
- Earliest claimed: 2008-09-24
- Assignees: NEC Laboratories America, Inc.
- Inventors: Tianbao Yang; Yun Chi; Shenghuo Zhu; Yihong Gong
- Web page: https://patentgrep.com/patent/US20100076913A1

## CPC

- G06Q 30/02: Marketing; Price estimation or determination; Fundraising
- G06N 7/01: Probabilistic graphical models, e.g. probabilistic networks
- G06Q 10/10: Office automation; Time management
- G06Q 10/42: Determination of affinities or common interests between users
- G06Q 10/44: Identification of trends within social networks, e.g. identification of trending topics
- G06Q 10/48: using social graphs
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS

## Abstract

Systems and methods are disclosed to find dynamic social networks by applying a dynamic stochastic block model to generate one or more dynamic social networks, wherein the model simultaneously captures communities and their evolutions, and inferring best-fit parameters for the dynamic stochastic model with online learning and offline learning.

## Claims

### Claim 1

A method to find dynamic social networks, comprising:
a. applying a dynamic stochastic block model to generate one or more dynamic social networks, wherein the model simultaneously captures communities and their evolutions, and
b. inferring best-fit parameters for the dynamic stochastic model with online learning and offline learning.

### Claim 2 (depends on 1)

The method of claim 1, comprising generating community membership assignments for all individuals in the social network over all time steps.

### Claim 3 (depends on 1)

The method of claim 1, comprising generating a collection of snapshot graphs for the dynamic social network.

### Claim 4 (depends on 1)

The method of claim 1, comprising applying the community assignments for an individual to track his or her community membership evolution over time.

### Claim 5 (depends on 1)

The method of claim 1, comprising applying the community assignments for all individuals at a given time step t to obtain an aggregated community structure at time

### Claim 6 (depends on 1)

The method of claim 1, comprising applying the community assignments for all individuals at all time steps to obtain the communities and their evolutions.

### Claim 7 (depends on 1)

The method of claim 1, comprising using either an offline or an online Bayesian inference learning algorithm to update the model.

### Claim 8 (depends on 1)

The method of claim 1, comprising applying Gibbs sampling or simulated annealing algorithm to learn the model parameters.

### Claim 9 (depends on 1)

The method of claim 1, comprising generating a dynamic stochastic model to model communities and their evolutions in a unified framework.

### Claim 10 (depends on 1)

The method of claim 1, comprising applying the dynamic stochastic block model and a Bayesian inference to a collection of snapshot graphs.

### Claim 11 (depends on 1)

The method of claim 1, comprising tracing and summarizing a community membership of a given individual.

### Claim 12 (depends on 1)

The method of claim 1, comprising aggregating the individual community memberships at a given time step t into the community structure of the social network at time t.

### Claim 13 (depends on 1)

The method of claim 1, comprising aggregating the individual community memberships over all time steps into the communities and their evolutions of the dynamic social network.

### Claim 14

A system to find dynamic social networks, comprising:
a. a unified framework with a dynamic stochastic block model to generate one or more dynamic social networks, wherein the model simultaneously captures communities and their evolutions, and
b. a Gibbs sampling module or a simulated annealing module to learn the model parameters and to infer best-fit parameters for the dynamic stochastic model with online learning and offline learning.

### Claim 15 (depends on 14)

The system of claim 14, comprising a module to generate community membership assignments for all individuals in the social network over all time steps.

### Claim 16 (depends on 14)

The system of claim 14, wherein the model generates a collection of snapshot graphs for the dynamic social network.

### Claim 17 (depends on 14)

The system of claim 14, wherein the community assignments for an individual is used to track his or her community membership evolution over time.

### Claim 18 (depends on 14)

The system of claim 14, comprising a module to apply the community assignments for all individuals at a given time step t to obtain the aggregated community structure at time t.

### Claim 19 (depends on 14)

The system of claim 14, wherein the community assignments for all individuals at all time steps are used to obtain the communities and their evolutions.

### Claim 20 (depends on 14)

The system of claim 14, comprising either an offline or an online Bayesian inference learning algorithm to update the model.
