جلال جواد

With the rapid expansion of social network usage, studying the influence of different nodes within these networks has become of great importance, especially those nodes that play a pivotal role in spreading information and shaping public opinion. This type of node is known as influencer nodes, which are nodes that have a high level of interaction and connection with the rest of the network. Over time, these nodes can change in terms of influence, increasing or decreasing in importance based on a set of factors such as the changes in the topics they discuss and the sentiments associated with them. This study proposes a comprehensive framework for predicting the evolution of influencer nodes based on topics and sentiments. Through a series of research steps, including data preprocessing, influential nodes were identified using centrality degree. The data were divided into "24-time windows" for temporal analysis, and dynamic graphical networks were generated that capture user interaction. The Topics and the Sentiments were extracted to track the evolution of influencer nodes, and predictive models were developed to predict future influencer node trends based on the interaction between topics and sentiments. We used a method to track the temporal evolution of influencer nodes in time series analysis. By integrating topic and sentiment domains, We used and eva‎luated "10 machine learning models". The results indicate that the "Random Forest" model outperformed other models, achieving superior performance in accuracy (92.31%), F1 scores (86%), and execution time (0.67 sec). These models contribute to reducing computational complexity and processing time, thus facilitating more efficient and informed decision-making in applications that require a comprehensive understanding of the dynamic nature of influencer nodes.

تكامل گره اينفلوئنسر، پيش‌بيني تكامل اينفلوئنسر، مدل‌هاي موضوعي، تحليل احساسات، جنگل تصادفي.

Influencer node evolution , Predicting the evolution of influencer , Topic models , Sentiments Analysis , Random Forest.

JALAL JAWAD

Dr. Behrouz Minaei