Research on Topic Recognition of Network Sensitive Information Based on SW-LDA Model

Abstract

The mining of network sensitive information is of great significance for understanding the social stability of the network. Obtaining the network public opinion of sensitive information is helpful to master Internet users' attitudes toward important social events. The related artificial intelligence technology can achieve the topics from the network texts. At present, the current topic recognition model has a low recognition rate for sensitive information and usually generates some inaccurate topic keywords. In this paper, a topic recognition method of the network sensitive information based on a sensitive word weighted-latent Dirichlet allocation (LDA) model is proposed. First, the basic sensitive word vocabulary is constructed by manual collection, and the embedding representation of the word is obtained through the training of a large amount of network corpus based on Word2vec. The semantic similarity between the word embedding is calculated to extend the basic sensitive word vocabulary. Second, the extended sensitive word vocabulary is embedded in the LDA model. On the one hand, it can improve the semantic understanding and the recognition ability of LDA for the sensitive topic words and promote the quality of the generated topic words. On the other hand, it can also improve the relevance of the topic keywords and the related topics and find more fine-grained keywords. The experimental results show that the sensitive word weighted-LDA model can effectively improve the topic recognition quantity and quality of sensitive information. This paper is helpful to the development of artificial intelligence. The generated corpus in this paper is meaningful to the research of text classification, clustering and information retrieval, and so on.INDEX TERMS Sensitive information, topic recognition, word embedding, artificial intelligence.

Idea Generation and New Direction for Exploitation Technologies of Coal-Seam Gas through Recombinative Innovation and Patent Analysis

“…The true value of patent analysis stems from its capability of describing the content of technology based on the relationships between keywords [42]. Some words that can reflect the technical characteristics of sensitive information appear in the text at a low frequency, which will cause most words that can represent the text to be drowned by some high-frequency words [43]. Many patent analyses employ k-means clustering [44][45][46].…”

“…In other words, texts on different topics may use different vocabularies [27]. LDA is widely used in many topic modelling methods and is the most typical topic model [43]. LDA is essentially a Bayesian thematic model, which has been widely used in recent years.…”

Topic Analysis of Microblog About “Didi Taxi” Based on K-means Algorithm

“…Mi Wenli [21] used probabilistic potential semantic analysis (PLSA) method for topic modeling of microblog data to find hot topics, which effectively solved the problem that K means algorithm was sensitive to the clustering center. Xu [22] proposed a topic recognition method for network sensitive information based on the weighted Latent Dirichlet Allocation (LDA) model. Experiments show that this method can effectively improve the quantity and quality of topic recognition of sensitive information.…”

Sensitive Information Topics-Based Sentiment Analysis Method for Big Data

“…In the ''Research on Topic Recognition of Network Sensitive Information Based on SW-LDA Model'' [3], the algorithms and experimental results related to the topic identification of sensitive information are introduced in detail. Based on the results of topic recognition of sensitive information in the article ''Research on Topic Recognition of Network Sensitive Information Based on SW-LDA Model'' [3], the research of sentiment analysis about these topics is conducted.…”