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

Dynamic social network representation learning method

A social network and learning method technology, applied in the direction of instruments, data processing applications, calculations, etc., can solve problems such as inability to evolve, difficult to accurately describe the evolution trend of nodes, and difficult to reflect important network attributes of social networks, so as to achieve the effect of enriching network information

Pending Publication Date: 2022-02-11
成都博智云创科技有限公司
View PDF5 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, traditional methods cannot measure various influencing factors in the network and build an accurate evolution model based on the basic mechanism of dynamic social network evolution, and it is difficult to accurately describe the different evolution trends of nodes in the network.
[0006] 2. It is difficult to reflect the important network attribute of social network - community structure

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
  • Dynamic social network representation learning method
  • Dynamic social network representation learning method
  • Dynamic social network representation learning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0058] A dynamic social network representation learning method is realized based on ternary closure pattern analysis and community structure; including the following steps:

[0059] Step 1: Construct a dynamic network sequence, based on a given set of time steps {1,2,...,T}, for the social network data of a certain time step, construct a subgraph G {a} =(V,E {a} ,W {a} ), a∈{1,2,…,T};

[0060] Among them, V is a set of nodes, representing users in the network; E {a} is the set of edges at the current time step, representing the relationship between users; W {a} is a set of weights, representing the connection strength between users; the dynamic network is a sequence of subgraphs of all time steps G={G (1) ,G (2) ,...,G (T)};

[0061] Step 2: Community division, according to the structure of the network, and mark the community where each node is located; since the dynamic network evolves with time, it is necessary to perform community division on the network at each time...

Embodiment 2

[0085] This example is four real social network datasets, including fb-messages, ia-facebook, ia-contacts and ia-retweet, the specific information of which is shown in Table 1.

[0086] Table 1 Statistics of Weibo-Douban Network Data

[0087]

[0088] Step 1: For each dataset, we first construct the corresponding dynamic network sequence G={G (1) ,G (2) ,...,G (T)}, and then learn the low-dimensional vector U={U of each node at different time steps by the present invention (1) , U (2) ,...,U (T)}.

[0089] Step 2: In order to verify the performance of the proposed method, this example selects five classic network representation learning methods for comparison, including: DeepWalk, LINE, Node2vec, DynamicTriad and TNE, where DeepWalk, LINE and Node2vec are static network representation learning models, and DynamicTriad and TNE are dynamic network representation learning models. After weighing the computational complexity and computational performance, we use dimension...

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 dynamic social network representation learning method, relates to the field of network representation learning, and solves the technical problem that an evolution mode of a real social network and a low-dimensional feature vector of a community structure cannot be kept during learning at present, wherein the method comprises ternary closure mode analysis and a community structure. According to the invention, after a ternary closure mode is analyzed, the evolution process of nodes and edges in a social network is jointly modeled by combining the influence of the nodes, the similarity of the nodes, the community structure and other important factors of the social network; according to the method, the network structure of the node and the low-dimensional representation of the dynamic feature learning node can be kept at the same time, and the learned low-dimensional representation is more discriminative by capturing different node evolution modes in the social network; and the community is one of the most important features of the real social network, and the learned network representation can well reflect the structure of the community, so that the user can be helped to obtain more useful information, and the ternary closure process is better optimized.

Description

technical field [0001] The invention relates to the field of network representation learning, in particular to a dynamic social network representation learning method. Background technique [0002] With the advent of the Internet age, many social network service platforms of various types have been born, which have greatly improved people's living standards. We also call this type of platform Online Social Network (OSN), such as Weibo, Douban, and Tieba in China, and Facebook, Twitter, and Instagram abroad. Researching and analyzing widespread social networks has high commercial and academic value, and has attracted a large number of domestic and foreign scholars to conduct research. For example, research on the problem of information diffusion in social networks can be applied to scenarios such as rumor detection, public opinion guidance, and influence maximization. However, the continuous accumulation of massive and rich data in today's online social networks has brought...

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
IPC IPC(8): G06Q50/00
CPCG06Q50/01
Inventor 王维成杨敏杨博
Owner 成都博智云创科技有限公司