Unlock instant, AI-driven research and patent intelligence for your innovation.

Community Discovery Method Based on Topic Interaction

A community discovery and theme technology, applied in the field of community discovery, can solve the problem of poor community description accuracy

Active Publication Date: 2018-07-10
NORTHWESTERN POLYTECHNICAL UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to overcome the shortcomings of existing community discovery methods for poor community description accuracy, the present invention provides a community discovery method based on topic interaction

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Community Discovery Method Based on Topic Interaction
  • Community Discovery Method Based on Topic Interaction
  • Community Discovery Method Based on Topic Interaction

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] refer to figure 1 . The specific steps of the community discovery method based on theme interaction in the present invention are as follows:

[0021] Step 1: Use LDA to extract the topic information of the text in the microblog network, and establish a hypergraph model according to each interaction information of users in the microblog network (ie posting microblogs and forwarding microblogs). The hypergraph model includes two types of topic nodes, user nodes and topic nodes; each publishing behavior constitutes a publishing edge u-t, where u represents a user, t represents a topic, and each forwarding behavior constitutes a forwarding edge u1-t1-u2, cum user u 1 from user u 2 forwarded the Weibo text with the subject t1.

[0022]

[0023] Step 3: After mapping the network into a hypergraph model, it is equivalent to discovering dense subgraphs in the hypergraph model. use Indicates the density of the subgraph V', where s represents all edges in the subgraph, ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a community discovery method based on topic interaction, which is used to solve the technical problem of poor community description accuracy in existing community discovery methods. The technical solution is to extract the subject information of texts, photos, etc. in the network within a certain period of time, and build a hypergraph model based on each interaction in the network. For the hypergraph model, calculate the weight of the hyperedge according to the user entropy and topic entropy, select the seed nodes in the user, and calculate the subgraph contribution of different nodes according to the hyperedge weight. Iteratively calculate the contribution of different nodes to obtain a dense subgraph. The dense subgraphs are aggregated to different degrees by a hierarchical clustering method to obtain communities at different levels. Since the initial subgraph is constructed starting from the seed node, and the subgraph contribution degree of different nodes is calculated according to the hyperedge weight, and then the contribution degree of different nodes is iteratively calculated to obtain a dense subgraph, and the dense subgraph is aggregated in different degrees to obtain The contribution value of different topic nodes to the community accurately describes the community.

Description

technical field [0001] The invention relates to a community discovery method, in particular to a topic interaction-based community discovery method. Background technique [0002] Social network is a kind of complex network and a heterogeneous network. The social network includes not only user nodes, but also topic nodes composed of text, location nodes composed of user check-in information, interest nodes composed of photos, etc., that is, social networks are no longer a large number of links of the same nature nodes, Rather, it is an interlinkage of many different types of nodes. The dense subgraph composed of different nature nodes due to the close interaction in the network is called the community in the network. [0003] There are two main methods for community discovery in complex networks: one is to simplify the complex network into a general network, that is, a network containing only nodes of the same nature, and then use user similarity to obtain communities in th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F17/30G06Q50/00G06K9/62
CPCG06F16/951G06Q50/01G06F18/295
Inventor 王柱於志文冯斌郭斌
Owner NORTHWESTERN POLYTECHNICAL UNIV