Machine Learning-Based Predictive Model for Service Quality Assessment and Policy Optimization in Adult Day Health Care Centers

Abstract

This paper presents a novel machine learning-based predictive model for service quality assessment and policy optimization in Adult Day Health Care (ADHC) centers. The proposed framework integrates Neural Network Boost (NNB) algorithms with cloud computing infrastructure to enhance service delivery efficiency and quality monitoring. The system architecture incorporates real-time health monitoring data from multiple sources, including IoT sensors and electronic health records, processed through a sophisticated data preprocessing pipeline. Experimental implementation across 15 ADHC centers, involving 2,854 elderly participants over a 12-month period, demonstrated significant improvements in service quality prediction accuracy. The NNB model achieved 94.3% accuracy in quality assessment, representing a 15.3% improvement over traditional methods. The policy optimization component, utilizing reinforcement learning techniques, generated a 28.5% improvement in resource utilization and 32.7% increase in service delivery efficiency. The system's real-time monitoring capabilities reduced manual evaluation time by 65%, enabling enhanced direct patient care. Comprehensive validation across multiple operational scenarios confirmed the model's robustness and scalability. The implementation results demonstrate the framework's effectiveness in addressing the complex challenges of ADHC service quality assessment and policy optimization, providing valuable insights for healthcare administrators and policy makers.