Integration of deep learning and knowledge graphs for intelligent agent decision optimisation

Enhancing the functionality of augmented reality using deep learning, semantic web and knowledge graphs: A review

Human–robot collaboration has gained attention in the field of manufacturing and assembly tasks, necessitating the development of adaptable and user-friendly forms of interaction. To address this demand, collaborative robots (cobots) have emerged as a viable solution. Deep Learning has played a pivotal role in enhancing robot capabilities and facilitating their perception and understanding of the environment. This study proposes the integration of cobots and Deep Learning to assist users in assembly tasks such as part handover and storage. The proposed system includes an object classification system to categorize and store assembly elements, a voice recognition system to classify user commands, and a hand-tracking system for close interaction. Tests were conducted for each isolated system and for the complete application as used by different individuals, yielding an average accuracy of 91.25%. The integration of Deep Learning into cobot applications has significant potential for transforming industries, including manufacturing, healthcare, and assistive technologies. This work serves as a proof of concept for the use of several neural networks and a cobot in a collaborative task, demonstrating communication between the systems and proposing an evaluation approach for individual and integrated systems.

General Background: The rapid advancement of digital technology and artificial intelligence has created new opportunities for enhancing learning quality, including within Islamic boarding schools (pesantren). Specific Background: Pesantren, as traditional Islamic institutions, increasingly face the need to integrate modern pedagogical approaches while safeguarding Islamic values. Knowledge Gap: Although research on technology-based education is expanding, studies focusing on the integration of deep learning, Islamic values, and digital technology specifically in pesantren remain limited. Aims: This study aims to conceptualize a deep learning model that harmonizes Islamic teachings, digital tools, and 21st-century competencies to strengthen both academic and character formation in pesantren. Results: Through a literature review, the study identifies three core dimensions of an integrated learning design: Islamic values as the moral-spiritual foundation, digital technology and AI as pedagogical enhancers, and 21st-century skills as essential competencies for santri. Novelty: This study proposes a holistic framework that positions deep learning not only as a cognitive process but as a spiritually grounded, technology-supported educational paradigm unique to pesantren. Implications: Successful implementation requires cultural adaptation, improved digital infrastructure, and continuous teacher training to ensure ethical and effective technology use. Highlights: Integration of Islamic values, digital tools, and deep learning creates a holistic pesantren learning model. Digital technology enhances pedagogy but requires cultural adaptation and teacher training. The framework prepares santri with both spiritual grounding and 21st-century skills. Keywords: Deep Learning, Pesantren, Islamic Values, Digital Technology, 21st-Century Competencies

Chapter 3 - Sensor-enabled biomedical decision support system using deep learning and fuzzy logic

This study examines the effectiveness of Hybrid Cultural Learning integrating Deep Learning and Problem-Based Learning (PBL) in enhancing students’ cultural literacy in an Educational Psychology course. A mixed-methods sequential explanatory design was employed with 72 students divided into experimental and control groups. Quantitative results indicate a significant improvement in cultural literacy in the experimental group (t = 7.842; p < 0.001), with mean scores increasing from 67.4 to 84.9, while the control group showed no significant change. Qualitative findings reveal that the integrated approach enhances contextual understanding, student participation, collaboration, and higher-order thinking skills (HOTS). Overall, the findings suggest that the integration of Hybrid Cultural Learning, Deep Learning, and PBL is effective in improving students’ cultural literacy in Educational Psychology learning.

This study explores the integration of the pedagogical Deep Learning (DL) approach, specifically the Meaningful, Mindful, and Joyful (MMJ) framework, within the traditional educational system of Indonesian pesantren. The aim is to analyze the compatibility, challenges, and operationalization of DL to enhance learning outcomes and prepare students for the demands of the digital era. A qualitative case study approach was employed at pesantren Darunnajah in Bogor, involving in-depth interviews with 12 participants (3 teachers, 6 students, and 3 administrators), observations, and document analysis. The findings highlight a strong theoretical alignment between DL and pesantren's core educational values, such as Fahm (deep comprehension), Ijtihad (critical reasoning), and Adab (moral development). Empirically, teachers reported improved student engagement and reflective learning practices, with one educator noting, "The MMJ framework helps students connect religious lessons to their daily lives, making learning more relevant." However, the study identifies key challenges, including infrastructure gaps, diverse levels of digital literacy, and the need to balance digital innovation with the preservation of moral guidance. Strategies such as blended learning models and focused teacher training are recommended. The integration of DL within pesantren represents a strategic pedagogical evolution, empowering students to develop both religious and modern competencies. Future research should focus on the longitudinal impacts of DL on student outcomes.

The integration of Deep Learning (DL) into Manufacturing Execution Systems (MES) and Enterprise Resource Planning (ERP) systems would greatly affect the development of smart manufacturing, therefore giving transformational powers for efficient resource allocation and real-time production optimisation. Data-driven insights, predictive maintenance, dynamic scheduling, and proactive decision-making help DL models including Recurrent Neural Networks (RNNs), Convolutional Neural Networks (CNNs), and Reinforcement Learning (RL) agents learn from enormous volumes of historical and real-time sensor data, uncover hidden patterns, and adapt to manufacturing environments unlike conventional rule-based approaches. By forecasting equipment failures and material shortages, DL improves MES response; at the same time, it guides ERP decisions including procurement, labour allocation, and energy control when integrated with MES, which controls shop-floor control, and ERP, which supervises enterprise-level planning. These models can close the loop between data collecting, analytics, and execution. Autoencoders and transformers also assist by mimicking challenging events and spotting system behaviour issues. With this combined DL-MES-ERP architecture real-time feedback and self-optimization is achievable using edge computing and cloud platforms for scalable, low-latency inference. The results show observable benefits in general equipment effectiveness (OEE), lower unplanned downtime, better inventory control, and more production agility using industrial simulations. Emphasising important DL techniques especially suited for smart factory use-cases, the paper investigates the performance of modular integration frameworks. Among the topics addressed are data heterogeneity, model generalisation, cybersecurity concerns, and human-machine collaboration. Ultimately, our work underlines the significance of DL in producing autonomous, adaptable, robust cyber-physical production systems resistant against increasing complexity and demand unpredictability. The integration of Deep Learning (DL) techniques with Manufacturing Execution Systems (MES) and Enterprise Resource Planning (ERP) systems is investigated in order to enhance real-time production optimisation and resource allocation in smart manufacturing environments. Data-driven intelligence obtained from networked systems allows manufacturers to achieve predictive capabilities, adaptive planning, and autonomous control. By offering an architecture for integration, evaluations of primary deep learning models relevant to manufacturing data, and results showing better use of resources, lower downtime, and improved production efficiency, the research proves the architecture using case studies and simulations.

A Price Risk Early Warning Model for Agricultural Products Based on the Integration of Deep Learning and Knowledge Graphs

Background: Magnetic Induction Tomography (MIT) faces significant challenges in imaging low-conductivity materials, particularly in optimizing multi-frequency excitation parameters for enhanced detection sensitivity. Conventional approaches face challenges in processing weak electromagnetic responses from low-conductivity materials (10^-18 to 10^-12 S/m). This limitation results in poor image quality and reduced detection capabilities. Purpose: This study introduces an innovative adaptive multi-frequency optimization framework integrated with deep learning for MIT, specifically designed to enhance the detection and characterization of low-conductivity materials. The framework introduces a novel HBDL-TVR-MF-ACC-MIT algorithm that dynamically optimizes excitation frequencies while leveraging deep learning for improved signal processing and image reconstruction. Method: We developed an integrated approach combining adaptive frequency optimization (1 kHz - 10 MHz) with deep learning architectures. The system employs frequency-hopping techniques and custom-designed CNN for optimization and reconstruction. The framework was validated through comprehensive COMSOL Multiphysics simulations and experimental testing using standardized phantoms. Results: The framework demonstrated substantial improvements in MIT imaging performance, including enhanced detection sensitivity for ultra-low conductivity materials, significant reduction in reconstruction time, and improved spatial resolution. The system achieved consistent performance across diverse material types, with notable improvements in image quality metrics and system stability. Key achievements include a 45% reduction in reconstruction time and 40% improvement in spatial resolution compared to conventional methods. Conclusion: This adaptive multi-frequency optimization approach represents a significant advancement in low-conductivity MIT imaging, enabling accurate and efficient detection of previously challenging materials. The integration of deep learning with optimized frequency selection establishes a robust framework for non-invasive imaging applications in medical diagnostics and industrial monitoring, with potential for substantial cost reduction and efficiency improvements in both sectors.

The advent of the fourth industrial revolution (industry 4.0) has sparked a transformation in manufacturing through the integration of deep learning (DL) and big data technologies. These innovations have enhanced the ability of manufacturers to process and analyze vast amount of data, providing insights that improve decision-making, efficiency, and overall productivity. Deep learning, a subset of artificial intelligence (AI), offers advanced capabilities in data pattern recognition and predictive modeling, while big data facilitates the management of large and complex datasets from various sources. However, literature on the integration of deep learning, big data and industry 4.0 is still limited in the manufacturing context. This paper provides a detailed overview of big data, deep learning and industry 4.0 in manufacturing. It explores the applications of deep learning and big data in manufacturing, highlighting their role in optimizing production processes, predictive maintenance, quality control, and supply chain management. Furthermore, the paper addresses the key challenges and solutions associated with the integration of these technologies, such as data privacy, security, computational complexity, and the need for skilled labor. KEYWORDS: Deep learning, Industry 4.0, Supply chain management, Big data, Predictive maintenance, Artificial intelligence, ERP systems.

The integration of Deep Learning with IoT technology is a significant advancement in the field of artificial intelligence. The chapter explores the integration of IoT architectures and deep learning frameworks, discussing important strategies for data collection, processing techniques, and deep learning model development, focusing on edge and cloud computing. The chapter showcases practical applications in smart engineering, including predictive maintenance, smart manufacturing, energy management, and environmental monitoring. Real-world case studies are presented to demonstrate the application of these technologies and address common challenges and solutions. The chapter predicts future trends in IoT and deep learning, highlighting emerging technologies and potential advancements in smart engineering, promising enhanced efficiency, predictive capabilities, and sustainability.

This study investigates the use of deep reinforcement learning techniques to improve traffic signal control systems through the integration of deep learning and reinforcement learning approaches. The purpose of a deep reinforcement learning architecture is to provide adaptive control via a reinforcement learning interface and deep learning for the representation of traffic queues with regards to signal timings. This has driven recent research, which has reported success in the use of such dynamic approaches. To further explore this success, we apply a deep reinforcement learning algorithm over a grid of 21 interconnected traffic signalized intersections and monitor its effectiveness. Unlike previous research, which often examined isolated or idealized scenarios, our model is applied to the real-world traffic network of Via “Prenestina” in eastern Rome. We utilize the Simulation of Urban MObility (SUMO) platform to simulate and test the model. This study has two main objectives: ensure the algorithm’s correct implementation in a real traffic network and assess its impact on public transportation, incorporating an additional priority reward for public transport. The simulation results confirm the model’s effectiveness in optimizing traffic signals and reducing delays for public transport.

The era of big data produces massive data, and carrying out data mining can effectively obtain effective information in huge data, which provides support for efficient decision-making and intelligent optimization. The purpose of this paper is to establish a digital twin system, preprocess massive data using random matrix theory, and design the knowledge graph construction process based on digital twin technology. The BERT model, attention mechanism, BiLSTM model, and conditional random field of the joint deep learning technology are used to identify the knowledge entities in the digital twin system, extract the knowledge relations through the Transformer model, and utilize the TransE model for the knowledge representation in order to construct the knowledge graph. Then, the constructed knowledge graph is combined with the multi-feature attention mechanism to build an anomaly data prediction model in the digital twin system. Finally, the effectiveness of the methods in this paper is validated through corresponding experiments. The TransE model is used for knowledge representation. The accuracy of ternary classification is higher than 80% in all cases, and the MR value decreases by up to 64 compared to the TransR model. The F1 composite score of the anomaly data prediction model is 0.911, and the AUC value of the validation of knowledge graph effectiveness is 0.702. Combining deep learning with the knowledge graph, the knowledge information can be realized in the digital twin system’s accurate representation and enhance the data mining ability of the digital twin system.

A novel approach for water quality classification based on the integration of deep learning and feature extraction techniques

With the rapid development of intelligent transportation, autonomous driving has become a core direction of technological innovation. Its safe and efficient operation relies on precise processing of multi-source sensor signals. Traditional signal processing methods, limited by manual feature design, often fail to extract complete features and suppress noise in complex traffic environments. This leads to perception biases and decision-making delays, posing safety hazards and restricting the implementation of autonomous driving technology. By combining literature review and technical analysis, this paper focuses on the integration of deep learning and autonomous driving signal processing, aiming to explore how to enhance the efficiency and reliability of signal processing through deep learning algorithms. It concludes that deep learning, through its adaptive feature learning ability, can significantly improve the real-time performance and accuracy of signal processing. However, challenges such as data scarcity, poor model interpretability, and high computational costs still exist. Future research should focus on technological breakthroughs such as lightweight model design and multi-sensor fusion to promote the large-scale application of deep learning in autonomous driving signal processing.