Integrating deep learning and big data analytics for enhanced smart city infrastructure and human resource management systems

In this paper, a framework has been proposed based on EfficientNet with big data analytics to solve the problems in smart city infrastructure and human resource management systems (HRMS). As a result of efficient performance on various datasets such as NYC Taxi Trip Records, IBM HR Analytics Employee Attrition Dataset, and London Smart Meters Dataset, we apply EfficientNet’s compound scaling and squeeze-and-excitation techniques with fast scalable big data pipelines to further enhance predictive performance. The proposed system gives results that clearly outperform five recent state-of-the-art methods with accuracy of 92.3%, a micro F1-score of 90.1%, and an MCC of 0.85 and exhibits its robust scalability with the growing dataset sizes. The methodology combines deep learning and big data analytics to improve predictive performance. The study spotlights its strengths and limitations (such as interpretability and computational costs) of deep learning methods and points out potential future directions including explainable AI and distributed training frameworks. By connecting theoretical innovation with practical implementation, this research provides an innovative solution for optimal dynamic multi-task assignment in urban environments realizing smart cities and urban settlements, and workforce management strategies for the modern world of data.