Online Public Opinion Communication Analysis: Natural Language Processing and Graph Analysis are Integrated

Source: "Information Security and Communication Confidentiality"

Author: Xu Ming, Wei Guofu, Yin Qian'an

"Artificial Intelligence Technology & Consulting" was released

Optimize identification and traceability capabilities

With the rapid development of network information technology, doing a good job in the analysis of network public opinion is the basic premise for ensuring the healthy and stable dissemination of network public opinion information.

Introduction

Figure 1 Model idea of combining natural language processing and graph analysis

Methodology

2.1 Keyword Screening

2.1.1 Preprocessing of original corpus data

2.1.2 Application of text tokenization

2.1.3 Screening of target keywords

Figure 2 Extraction method for text corpus keywords

2.2 Topic sentence extraction

Figure 3 The extraction process of a text corpus subject sentence

2.2.1 Word embedding processing

2.2.2 Construct an undirected authority graph

2.2.3 Topic sentence extraction

2.3 Analysis of public opinion communication

2.3.1 Construct a relationship graph of public opinion topics

2.3.2 Analysis of public opinion communication

Figure 4 An approach to group segmentation of complex network structures

Figure 5 Graph analysis method in the analysis of public opinion topics is shown at some nodes

Table 1 Comparison between the semantic analysis of the bidirectional LSTM+CRF model and the traditional analysis

Figure 6 Schematic diagram of the traceability of the path of public opinion propagation

Disclaimer: The articles and pictures reproduced by the official account are for non-commercial educational and scientific research purposes for everyone's reference and discussion, which does not mean that they support their views or confirm the authenticity of their content. The copyright belongs to the original author, if the reprinted manuscript involves copyright and other issues, please contact us immediately to delete.

"Artificial Intelligence Technology & Consulting" was released