Hybrid Deep Learning Modelfor Enhancing the Streaming Efficiency of 6G Enabled Massive IoT Systems

6G networks are anticipated to provide a wider range of capabilities compared to previous generations, potentially accommodating applications beyond existing mobile apps, including virtual and augmented reality, artificial intelligence (AI), and the Internet of Things (IoT). IoT data analytics is complex, requiring more supporting processes to provide high accuracy.The raw data generated from IoT systems is not certain, and it cannot be processed due to many variations, outliers, missing elements, wrong and unconditional data flow, mismatched data types, and non-defined data size. The massive amounts of data generated by 6G-enabled IoT systems make artificial intelligence (AI) and machine learning (ML) crucial in reconfiguring and improving their performance. Thus, recent research focuses on pre-processingto improve data quality and use learning models for prediction. It increases the computational and time complexity of the whole work. This problem is considered the major problem, and this research implements Baye’s Theorem for predicting hypothesis-based data to be forecasted and streamed from one place to another, which has less complexity and takes less time. AI and ML are crucial in enhancing and improving IoT systems provided by 6G because of the vast quantity of exploratory data produced. This researchuses a hybrid deep learning strategy to improve the streaming efficiency of 6G Enabled Massive IoT Systems. The proposed EDA and Baye’s Theorems are implemented in Python, and the results are verified.The results are compared with other similar methods to evaluate its performance regarding prediction accuracy and streaming efficiency. The overall experiment shows that the data streaming process obtained using EDA-Baye’s Theorem outperforms the other methods with high efficiency.