In today’s data-driven world, big data environments are becoming increasingly complex, characterized by high volume, variety, and velocity. Traditional data processing methods are no longer sufficient to handle such challenges. Artificial Intelligence (AI) provides powerful solutions for extracting value from diverse and dynamic data sources. This paper reviews key AI techniques—including machine learning, deep learning, natural language processing, graph-based models, and federated learning—and discusses their applications in complex scenarios such as healthcare, finance, smart cities, and Industry 4.0. It also highlights major challenges, including data quality, model interpretability, computational cost, and privacy concerns. Finally, the paper explores future directions in AI development, such as multimodal learning and real-time decision-making. These advancements will play a vital role in enabling intelligent, efficient, and ethical data analytics in the years to come.

Show more

Traditional methods, such as polygraphs, suffer from limitations including single-modality vulnerability, cultural bias, and adversarial attacks. This paper presents a novel Multimodal Physiological-Behavioral Fusion System (MPBFS) that integrates seven modalities: microexpressions, speech, text, eye movements, galvanic skin response (GSR), cultural features, and adversarial defense mechanisms.The system employs a Cultural Dimension-Physiological Signal Cross-Validation Module to dynamically weight modalities, reducing cultural bias.The Cascade Lightweight for Edge Computing achieves 84.6% accuracy (F1=0.87) on cross-cultural datasets with 23 FPS. A Two-Channel Adversarial Defense: Detects DeepFake audio via phase discontinuity analysis and validates microexpressions using Lagrangian biomechanical modeling confirming enhanced cross-cultural robustness. Experiments demonstrate a 31% improvement in Cultural Stability Index and a 42% reduction in adversarial attack success rates, validating the system’s robustness. Designed for forensic interrogation and security screening, MPBFS integrates dynamic anonymization and cross-modal validation to ensure ethical AI deployment, addressing key limitations of traditional deception detection methods.

Show more

With the rapid development of Internet of Vehicles (IoV) technologies, the automotive industry is undergoing a profound transformation from traditional mechanical systems to intelligent and connected systems. By integrating technologies such as 5G communication, V2X, edge computing, and artificial intelligence, the IoV has established an intelligent transportation system characterized by vehicle-road-cloud-terminal collaboration. Focusing on the two core topics of automotive safety and user experience in the IoV era, this paper systematically reviews the security challenges and pathways for optimizing user experience, while also exploring future trends in their coordinated development. The study finds that IoV technologies offer significant advantages in enhancing traffic safety, improving traffic flow, and reducing energy consumption. However, they also face risks such as cybersecurity threats, data privacy breaches, and system reliability issues. Measures such as optimizing smart cockpit interaction and expanding full-scenario service ecosystems can effectively enhance both automotive safety and user experience. In the future, as technology continues to advance and supportive policies are further implemented, IoV technologies will drive the automotive industry toward a safer, smarter, and more efficient direction.

Show more

With the rapid development and widespread popularity of the Internet, the amount of data in social media and networks is growing exponentially, and sentiment analysis for this huge amount of data is very complex but significant. Fine-grained sentiment analysis has become the choice of researchers when dealing with various sentiment analysis tasks. Different from coarse-grained sentiment analysis, which only focuses on emotional polarity, fine-grained sentiment analysis involves emotional polarity and emotional intensity and recipients, providing more specific information about emotions. This paper aims to provide relevant research methods on fine-grained sentiment analysis and apply them to social network texts to analyze the challenges and solutions. This paper will classify fine-grained sentiment analysis from three methods: rule-based, machine learning and deep learning. This research finds that fine-grained sentiment analysis can not only accurately capture the emotions in the text, but also judge the direction and intensity of emotions, and understand different types of emotions in the text more specifically. This is of great help in dealing with more complex texts, such as social network texts. Combining fine-grained sentiment analysis with various large models can solve many challenges and problems when dealing with social network texts.

Show more