Predicting Social Media Behavior with a Fuzzy RPD Algorithm: A Hybrid Approach with Machine Learning Accuracy

This study introduces an enhanced fuzzy RPD model to analyze user behavior on social media platforms such as Twitter, Facebook and Instagram. Extending the statistical framework developed by DeLegge and Wangler, we formulate a system of ordinary differential equations representing three distinct users' categories: Receptive, Pathogenic and Detached. Key model parameters social media advertising and withdrawal rates are treated as fuzzy variables influenced by trending strain, modeled using triangular fuzzy numbers. Numerical simulations are conducted using Eulers method over a month of period. A critical highlight of this work is exponential model accuracy, validated by performance metrices such as MSE, RMSE, MAE and MAPE for Receptive, Pathogenic and Detached populations, all of which indicate minimal error. These results confirm the robustness and reliability of the simulation