Social Media Analysis Through Sentiment and Influential Node Analysis

In today's digital environment, the growth of social media has created unprecedented challenges and opportunities for organizations to understand and influence public opinion.This study offers an integrated approach to social media that combines advanced theory with empirical research to determine the complexity of online relationships.Leveraging natural language processing (NLP) technology and advanced network analysis, our program provides a powerful tool to visualize social media trends and identify key individuals creating narratives.Based on a fine-tuned "distillbert" transformer model, our analysis theory can classify content as positive, negative, or neutral with an F1 score of 0.86, indicating its ability to accurately capture human emotions. .At the same time, our influence research uses network analysis to identify individuals or landmarks important to the spread of information and influencers who can influence public opinion.In addition to its analytical capabilities, the project provides important information to promote online relationships and reduce the prevalence of negative emotions, setting an example for ethical participation in the digital world.In summary, our combined approach represents a significant advance in social analytics, allowing organizations to work efficiently and engage with digital users in the rapidly changing environment of pace of social relations.