Analysis Method of Sina Weibo User's Emotional Influence Based on Deep Learning

Abstract

The present invention provides a method for analyzing the emotional influence of Sina Weibo users based on deep learning, including data acquisition, data preprocessing, constructing multiple emotional feature word vectors, text sentiment classification and defining the emotional influence of Weibo users. In the present invention Using the MSF-CNN algorithm to perform sentiment analysis on Weibo texts is superior to the w2c-CNN algorithm, which can well cooperate with the language environment of free writing and random words in Weibo, and can well consider the context of Weibo texts environment, extract text features, and consider the emotions carried by the words themselves, which can better classify emotions.

Description

technical field [0001] The invention relates to the field of data algorithms, and in particular to a method for analyzing the emotional influence of Sina Weibo users based on deep learning. Background technique [0002] Algorithm refers to an accurate and complete description of a problem-solving plan, and is a series of clear instructions to solve a problem. Algorithm represents a systematic method to describe the strategy mechanism for solving problems, that is, to a certain standard input, within a limited time. Obtain the desired output. If an algorithm is flawed, or is not suitable for a problem, executing this algorithm will not solve the problem. Different algorithms may accomplish the same task with different time, space, or efficiency. One algorithm The pros and cons can be measured by space complexity and time complexity. It has been more than ten years since we entered the Web 2.0 era. The rapid development of Internet technology and the rapid popularization of mo...

AI Technical Summary

Problems solved by technology

[0003] However, since the current research focus of influence analysis is mainly on the topological structure and interactive information of social networks, this research method is not considered comprehensively, because in social network platforms such as Weibo, users often publish some content with Emotionally inclined blog posts are used to capture the interest of others, thereby interacting with others. This shows that the information of Weibo blog posts is also a key factor in measuring user influence, but this aspect has been ignored by many researchers.

Images