Computational intelligence in cyber-physical systems and real-world big data analysis

The paper presents a Distinctive Functional Analytics Model (DFAM) for managing big data in Cyber-Physical Systems (CPS) that tries to make analytics on data more effective and accurate. CPS integrates computational, communication, and hardware systems to couple physical and internet platforms for real-time data processing. DFAM addresses processing large, composite data through self-trained smart learning to handle data integrity and normality with high-precision outputs and error minimization through stochastic computing. The model is reusable and flexible in real-world applications and guarantees output without interrupting computation. The results verify that DFAM outperforms other models by 7.42% more accuracy, 18.39% less latency, and 25.85% and 9.76% less complexity and error, respectively. Results verify the usability of the model to process actual-time large amounts of data on CPS platforms with an established solution for data-oriented applications and services.