Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Social message propagation range prediction method without topological structure

A technology of topology structure and prediction method, which is applied in the field of prediction of the spread range of social information, can solve problems such as problems that do not consider the mutual influence of information, and achieve the effect of improving accuracy and accurate prediction results

Active Publication Date: 2020-04-10
上海帮赋成科技有限公司
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0016] The purpose of the present invention is to solve the problem that the existing methods do not consider the mutual influence of messages during the propagation process, and propose a topology-free social message propagation range prediction method

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
  • Social message propagation range prediction method without topological structure
  • Social message propagation range prediction method without topological structure
  • Social message propagation range prediction method without topological structure

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0044] Specific implementation mode 1: In this implementation mode, a method for predicting the spread range of social information without a topology structure is as follows:

[0045] The invention studies the problem of predicting the range of message propagation under the condition of no topology structure, and proposes a method NT-EP for predicting the range of message propagation without topology structure. The method consists of four parts: (1) Construct a weighted propagation graph for each message by using the characteristics of message propagation decaying with time, use random walk strategy to obtain multiple propagation paths on the propagation graph, and then use the word2vec method to calculate each (2) replace the propagation path of the target message with the user’s feature vector sequence and input it to the Bi-Gate Controlled Recurrent Neural Network (Bi-GRU), and combine the attention mechanism to calculate the propagation feature vector of the target message;...

specific Embodiment approach 2

[0057] Specific embodiment 2: The difference between this embodiment and specific embodiment 1 is that in step 1, a weighted propagation graph is constructed for each message according to the propagation time difference in the message action log, as shown in figure 1 Shown in (a) and (b). The numbers on the side of the propagation graph represent the influence probability between users. After the propagation graph is constructed, use random walk to extract several possible propagation paths of the message from the propagation graph, such as figure 1 Shown in (c); the specific process is:

[0058] Transmission path selection

[0059] A given action log usually sorts the actions of each message according to the propagation time, such as figure 2 Shown in (a). User V 1 Received message A at time 1 1 , user V 2 Received message A at time 2 1 ,……. The real propagation track of the message cannot be obtained from the given action log. Because the real situation may be: use...

specific Embodiment approach 3

[0068] Specific embodiment three: the difference between this embodiment and specific embodiment one or two is that the word2vec method is used in the step two to calculate the initial feature vector of each target user on the propagation path of the message; the specific process is:

[0069] After extracting the propagation paths of all messages, treat each propagation path as a sentence, and each user on the path as a word in the sentence, input to word2vec [23] (Le Q,Mikolov T.Distributed representations of sentences and documents[C] / / International conference on machine learning.2014:1188-1196.) In the skip-gram model, the initial feature vector of each target user is obtained; assuming the user's initial The feature vector has dimension H.

[0070] Other steps and parameters are the same as those in Embodiment 1 or Embodiment 2.

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 social message propagation range prediction method without a topological structure, and relates to a social message propagation range prediction method. The objective of theinvention is to solve the problem that mutual influence of messages in a propagation process is not considered in an existing method. The method comprises the following steps of: 1, extracting a plurality of possible propagation paths of a message from a propagation graph by using a random walk mode after the propagation graph is constructed; 2, calculating an initial feature vector of each targetuser on a message propagation path by using a word2vec method; 3, obtaining a final vector representation of each target user on the propagation path; 4, calculating a propagation feature vector of each target message; 5, calculating influence vectors of other messages; and 6, combining the propagation feature vector of the target message obtained in the step 4 with influence vectors of other messages, and fitting an incremental propagation range of the target message by using MLP. The method is applied to the field of message propagation range prediction.

Description

technical field [0001] The invention relates to a method for predicting the spread range of social information. Background technique [0002] With the rapid development of social networks in recent years, more and more users use Sina Weibo, Twitter, Facebook and other social networking sites to share their lives. According to statistics, Facebook has more than 2.3 billion monthly active users as of December 31, 2018 [1] (Zephoria. The top 20 valuable Facebook statistics-up-dated April 2018. [Online], Available: https: / / zephoria.com / top-15-valuable-facebook-statistics / , January 1, 2019.). It can be seen that social networking has become a part of many people's lives. At the same time, major social platforms are also promoting the rapid dissemination of various news. For example, on Sina Weibo, hundreds of millions of microblogs are generated on average every day. A lot of important information will be contained in the Weibo generated every day. A user's update of a Weibo...

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): H04L12/24H04L12/58G06Q50/00G06Q10/04
CPCH04L41/147G06Q10/04G06Q50/01H04L51/52
Inventor 刘勇刘子图李晓坤
Owner 上海帮赋成科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com