The paper discusses how AI can be used to predict epidemics and improve public health responses on a wide range of critical topics from disease prediction using data to policymaking. Even conventional epidemiological models, often constrained by parameters, find it difficult to adapt to rapidly evolving disease dynamics. Our method combines machine learning (ML) and deep learning (DL) algorithms, such as long short-term memory (LSTM) and reinforcement learning (RL), to dynamically anticipate infection peaks and outbreak hotspots. Using both time-series and spatial data, the hybrid CNN-LSTM model predicted high-risk areas at a prediction accuracy of 89% and drastically improved the public health response planning. Further, the RL-based policy optimization system outperformed the traditional approach, enabling adaptive lockdowns and resource optimisation that reduced peak infection rates by 25% in dense regions. What we’ve found demonstrates AI’s ability to deliver both practical and predictive insights on epidemic dynamics and enable customisable, adaptive health policies. The work aims to advance AI-enabled epidemic management by offering a revolving door for predictive modeling and policy adaptations, responsive to changing epidemiological needs.

This paper discusses how big data analytics could revolutionize music education by suggesting pedagogic ideas and shaping learning experiences based on student profiles. Through a combination of practice records, teacher-student dialogue and qualitative feedback, this study builds detailed portraits of music students to inform adaptive, data-informed teaching. It includes collecting, processing and analyzing the data, and applying clustering to classify the learner into Fast Learners, Methodical Learners and Emotionally Driven Learners. Each profile gets personalized practice schedules, feedback, and incentives for improved engagement and skill acquisition. Findings demonstrate that personalized methods deliver considerable improvements in learning and achievement. It also uses adaptive learning pathways and gamification to enhance student progress. Adapting to the new realities through feedback loops and constantly updated adjustments, this big data approach holds the promise of a fresh and flexible music education model. This study not only suggests how big data can drive personalized learning but offers a template for future educational data mining in music and beyond.

This paper presents an in-depth learning process for Activities of Daily Living (ADL) and self-care planning in Adult Day Health Centers. The framework integrates multi-modal sensor data fusion with deep learning architectures to provide continuous monitoring and automated evaluation of older people's legal status. The system uses a hierarchical system combined with convolutional neural networks (CNNs) and long short-term memory (LSTM) networks, enhanced by monitoring systems for physical-spatial learning. Many data streams from wearable devices, environmental sensors, and medical monitoring devices are becoming pre-processed and de-processed. The framework incorporates a dynamic care plan adaptation strategy utilizing reinforcement learning techniques for intervention optimization. Experimental validation conducted across three Adult Day Health Centers with 150 participants over six months demonstrated superior performance compared to traditional assessment methods. The system achieved 92.8% accuracy in ADL recognition tasks, with a 35% reduction in assessment time and a 62% decrease in false alarm rates. Clinical validation through 25 detailed case studies revealed early detection of health deterioration, averaging 3.2 days ahead of conventional methods. The proposed framework significantly enhances the efficiency and accuracy of elderly care delivery while reducing healthcare provider workload by 40%. This research contributes to advancing intelligent healthcare by creating a solution for ADL measurement and self-correction.

The rapid innovation of Artificial Intelligence (AI) has transformed various sectors of society by revolutionising decision making and enhancing efficiency through novel data-driven technologies. This paper explores the challenges of striking a healthy balance between future AI innovation and personal data privacy, where the massive collection and utilisation of personal data have given rise to significant privacy concerns. The study identifies the risks of massive data collection, complex and opaque algorithms, and cybersecurity threats, while simultaneously highlighting the existing legal frameworks such as General Data Protection Regulation (GDPR) and the variations among global approaches to data privacy. The paper also discusses the technical solutions such as privacy-preserving techniques including differential privacy and federated learning, as well as encryption technologies that can facilitate the secure storage and transmission of data. The research proposes strategies for building privacy-preserving AI models and encouraging cross-industry collaboration to achieve a balance between innovation and the protection of individual privacy. It also adds to the ongoing discourse on shaping a responsible future for AI.

The paper discusses the use of predictive analytics and machine learning techniques like the LSTM neural network in order to enhance supply chain effectiveness. Economic Order Quantity (EOQ), JIT, and linear programming are well known for their supply chain optimization approaches, but they rarely take the challenge of an evolving supply chain into account. Such methods ignore real-time information and the non-linear dynamics of supply chain variables. Instead, the LSTM model takes historical and real-time data and calculates demand with high precision, making it easier to manage your inventory, avoid stockouts, and improve customer experience. As shown in the paper, LSTM provides an accuracy rate of 91% for forecasting demand, which is superior to the use of other conventional statistical methods like Moving Averages and Exponential Smoothing. The impact of this advanced machine learning technology for optimizing supply chain performance is huge, it reduces the operating cost, optimizes the use of resources, and delivers the best service. This work demonstrates how predictive analytics and machine learning can enable supply chain management to be flexible, effective and adaptable to market shifts.

This paper analyzes the efficacy of deep learning models, in particular reinforcement learning, in optimising the order execution for HFT. Typical execution models such as volume-weighted average price (VWAP) and time-weighted average price (TWAP) are typically unsuitable for HFT as they lack flexibility and are also latency. To solve these problems, our work utilizes a PPO reinforcement learning algorithm. Based on high-frequency tick data, the model continuously adapts to market conditions and enables performance enhancements in execution quality, speed and versatility under volatile markets. The experiment showed that the deep learning model performed better than the VWAP baseline, filling at 92%, market impact reduced by 15%, and running 43% faster. As this work shows, adaptive, deep learning based strategies not only enhance HFT execution, but also promote stability in the market, highlighting the possibilities of AI as a practical tool for trading.

The combination of Artificial Intelligence (AI) and Digital Twin technology provides disruptive potential to increase energy efficiency and carbon footprint in smart city infrastructure. Digital Twins — virtual copies of the real-world systems — are augmented by AI algorithms that enable continuous monitoring, predictive analysis and optimization. In this paper, we explore the use of AI-based Digital Twins on smart buildings, transport networks and smart grids to save significant amounts of energy and drive sustainability. This is done through machine learning and reinforcement learning algorithms which identify patterns of energy use with high precision and helps to reduce the energy usage in smart buildings by 25-30%. For transportation, AI-enabled traffic infrastructure reduced carbon emissions by 20% and enhanced EV infrastructure efficiency by 18%. The smart grids were better served by predictive energy distribution, which allowed for a 15% decrease in losses and a 20% rise in the use of renewable energy. All of these results point towards the potential of AI-augmented Digital Twins to reshape city planning, optimise resource consumption and play a key role in achieving global sustainability targets. This research underscores the need to embrace high-tech solutions for the next smart city projects to combat climate change and promote sustainable development.

Principal Component Analysis (PCA) is one of the most widely used dimensionality reduction methods in data analysis, which is renowned for its ability to handle the underlying structure of datasets. However, in the era of big data, characterized by high-dimensional, large-scale, noisy, and dynamic datasets, traditional PCA faces significant limitations. This paper reviews the challenges faced by PCA in big data environments and explores key extensions developed to enhance its applicability. Beginning with an overview of PCA’s mathematical principles, the paper identifies its inefficiency in dealing with massive datasets, data noise, and difficulties when applied to real-time environments. To solve these problems, various extensions of PCA have been created, including Incremental PCA, Sparse PCA, Kernel PCA, and Robust PCA. This survey further discusses practical applications of PCA in big data domains, including biological analysis, financial analysis, and image processing. Besides, the survey also examines the future directions of PCA research, such as combining PCA with advanced machine learning models, utilizing quantum computing to enhance efficiency, and ensuring privacy in PCA applications. This review aims to deepen the understanding of PCA in big data analysis, address the challenges, and reveal innovative solutions to enhance its efficiency and capability in handling high-dimensional and complex datasets for big data.

This paper proposes an enhanced Long Short-Term Memory (LSTM) based forecasting model explicitly designed for functional beverage sales prediction in cross-cultural markets. Traditional forecasting methods often struggle with capturing the complex interplay between cultural factors and sales patterns, leading to suboptimal prediction accuracy in diverse market environments. The proposed model incorporates innovative architectural modifications to the standard LSTM structure, integrating cultural-aware gates and specialized feature engineering techniques to capture market-specific characteristics. The research utilizes comprehensive sales data from six major markets across North America, Asia, and Europe from 2019 to 2023. The enhanced model demonstrates superior performance with a 28.4% improvement in prediction accuracy compared to traditional methods, achieving an average RMSE of 0.134 across all tested markets. The model's effectiveness is particularly evident in markets with high cultural diversity, where it achieved a 31.5% reduction in prediction error compared to conventional approaches. The research findings establish that cultural dimensions account for approximately 37.5% of sales variation across markets, highlighting the critical importance of cultural feature integration in sales forecasting. The practical implementation of the model resulted in a 23.7% reduction in inventory holding costs and improved resource allocation efficiency. This research contributes to the theoretical understanding of cross-cultural market dynamics and practical applications in global business operations.

At present, artificial intelligence has emerged as a significant field all over the world, offering diverse enhancements across many industries when combined with it. This study primarily examines the combination of the Internet of Things and artificial intelligence, highlighting practical applications and areas requiring enhancement. Mainly from a literature review utilising the three keywords: artificial intelligence, the Internet of Things, and application for search and screening, it was discovered that there are several issues in the application of this field. There are issues with delayed data transmission, compatibility, and anomaly detection in the field of autonomous driving. In the field of wearable devices, there are data security issues. In the field of industrial Internet of Things, there are problems such as low transmission efficiency, high latency, and inability to effectively resist external network attacks. In addition, there are also trust issues in the field of artificial intelligence where decisions and predictions cannot be understood by people. In besides this, this study provides several of recommendations for improvements in further research.