Community discovery method based on theme 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: 2015-12-16
NORTHWESTERN POLYTECHNICAL UNIV
View PDF3 Cites 6 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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0019] refer to figure 1 . The specific steps of the community discovery method based on theme interaction in the present invention are as follows:
[0020] 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.
[0021] Step 2: Calculate the user entropy of different users in U (set of users) related to different topics, and simultaneously calculate the topic entropy of different topics in T (set of topics) related to different users. The smaller the use...
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
Login to View More Abstract
The invention discloses a community discovery method based on theme interaction, and is used for solving the technical problem that a traditional community discovery method is poor in community description accuracy. The community discovery method has a technical scheme that theme information including texts, pictures and the like in a network within a time period is extracted, and a hypergraph model is established according to each-time interaction in the network. For the hypergraph model, the weight of a hyperedge is calculated according to user entropy and theme entropy, a seed node in a user is selected, and the subgraph contribution degrees of different nodes are calculated according to the weight of the hyperedge. The contribution degrees of different nodes are subjected to iterative computation to obtain a dense subgraph. The dense subgraph is subjected to clustering of different degrees by a hierarchical clustering method to obtain communities under different levels. Since an initial subgraph is constructed from the seed node and the subgraph contribution degrees of different nodes are calculated according to the weight of the hyperedge, the contribution degrees of different nodes are subjected to the iterative computation to obtain the dense subgraph, and the dense subgraph is subjected to the clustering of different degrees to obtain the contribution values of different theme nodes on the community and accurately describe 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
Login to View More Patent Type & Authority Applications(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