Artificial immune system algorithm with neural network approach for social media performance metrics

Social media plays a predominant role in this new era. Hence, the performance metrics of social media are usually evaluated by using statistical methods and machine learning approach. The core impetus of this research is to formulate a robust classification paradigm by integrating metaheuristic and neural network as a single paradigm. In this paper, artificial immune system algorithm incorporated with Hopfield neural network (HFB-3SATAIS) is proposed to evaluate the Facebook performance metrics data set. The proposed approach outperforms the standard Hopfield neural network with exhaustive search (HFB-3SATES) in terms of root mean square error (RMSE), mean absolute error (MAE), sum of squared error (SSE), symmetric mean absolute percentage error (SMAPE), Bayesian information criterion (BIC) and the CPU time. Thus, the accuracy, stability and robustness of the proposed network demonstrated a solid performance compared to the standard model. In addition, the work can be expanded by evaluating different class of benchmark data set such as time series data set.