Social user generation method based on emotional tendency analysis

Abstract

The invention discloses a social user generation method based on emotional tendency analysis. Existing user label generation methods do not consider emotional tendency of users to topics. The method comprises the following steps: on the basis of microblog content, constructing a semantic ontology by adopting a clustering algorithm, acquiring the microblog content of a user to perform sentiment classification, processing microblog information with sentiment tendency marks to obtain a bag-of-words model, extracting words of the bag-of-words model, acquiring category information by utilizing thesemantic ontology, and counting the category information with the sentiment marks; and adding a secondary label to obtain a final label, mapping the demographic information to the corresponding label,constructing a network graph, and generating the user in a semi-supervised manner according to an edge weight between the nodes of the network graph. According to the method, problems that the current social network users cannot automatically generate the users and the generated user tags are inaccurate are solved, the social users conforming to a certain theme can be generated, and a foundationis laid for subsequent social engineering research.

Description

technical field [0001] The present invention relates to the field of computer application and Internet technology, in particular to a user generation method based on social network text sentiment analysis. Background technique [0002] With the rapid development of Internet technology in the 21st century, users can publish information through social platforms in various ways. They can use text messages, emails, real-time software and web pages to publish their own edited information. Data associated with personal privacy. By researching and analyzing these data, interest tags and corresponding emotional tendencies related to users can be mined. [0003] The rise of social media in recent years has changed the way people live. People are keen to pay attention to their interests, share their preferences, and comment on popular trends on social networks, which also brings great value to data mining technology, which can be mined for a lot of instant, potential, and valuable k...

AI Technical Summary

Problems solved by technology

In terms of label construction, the traditional method builds a bag-of-words model from users’ tweets, maps them to the categories of the semantic ontology, and collects label information. However, this method can only be judged by the number of keywords, and does not consider the user’s interest in the topic. emotional tendency

Due to the existence of negative emotions, it will have an impact on some topic hobbies, thus interfering with the integrity of the generated users and greatly misleading the subsequent user analysis

Images