Social network influence based on SHIR and SLPR propagation models

The rapid progress of science and technology has revolutionized the dissemination of information, with the Internet playing a crucial role. While it has enhanced the ease of sharing information, it has also hastened the transmission of emotions on social media, sometimes resulting in unintended negative outcomes. This research seeks to tackle this issue by suggesting an innovative technique for analyzing social network influence using the Susceptible Hesitant Infected Removed (SHIR) and Susceptible Latent Propagative Removal (SLPR) propagation models. Through the development of an emotional communication model, we take into account the effects of news and public opinion on the rate of emotional communication among individuals. Furthermore, we investigate the impact of various network structures on user behavior. The findings from experiments demonstrate a notable relationship between changes in the density of emotion spreaders and hesitants and the influence of nodes in different network configurations. Specifically, the analysis reveals that the peaks of hesitators and disseminators were lower when the node influence was reduced. Additionally, we verified the precision and dependability of our model by examining data from the Baidu Index, a tool for big data analysis. The margin of error between the model and the actual data was minimal, underscoring the efficacy of our approach. In essence, the study highlights a direct correlation between the speed and extent of emotional propagation in social networks and the degree of nodes. The results showed that the density changes of emotion spreaders and hesitants were significantly correlated with the influence of nodes in different network settings. In the case of node influence of 0.86, the highest peaks of hesitator H and disseminator I were 0.101 and 0.109 lower than those of influence of 1.25. The data analysis of the Baidu Index showed that the maximum peak error of the model was only 0.04, which verified the accuracy and reliability of the model. This investigation carries significant implications for efficiently managing and steering the dissemination of emotions on social media, thereby promoting a healthier online environment.