Future Technology

The impact of AI-driven industrial upgrading on economic development

Xiaofei Hao, Aotip Ratniyom, Sivalap Sukpaiboonwat — Wed, 25 Jun 2025 00:00:00 +0000

The paper clarifies the interdependencies between AI adoption, industry upgrading, and economic development in the context of global digital transformation. With mixed-methods integrating econometrics and case studies, we test models formalizing mediating and threshold effects in AI-industry-economy relations. Our approach leverages a novel AI penetration score by industries alongside economic indicators and measures of industry sophistication. The results indicate that AI uptake mediates the pass-through of industry structure change to economic performance, with contribution levels increasing above certain thresholds. Evidence suggests that the association between the working-age population and economic growth varies by alternative industry upgrading rankings, with technologically sophisticated structures making better use of demographic opportunities. Threshold analysis identifies regimes where AI substitutes for traditional economic relations, revealing policy intervention points. These findings contribute to growth theory innovation by measuring AI's catalytic economic function and offer methodological innovation in the analysis of technological contributions. Strategic AI development agendas, human capital policies, and coordination mechanisms are among the key implications required to achieve inclusive growth in the digital economy. This study closes knowledge gaps on how demographic and technological drivers interact through industry structures to determine economic trajectories. Empirical results show that AI adoption mediates 52.8% of manufacturing sophistication's impact on GDP growth, threshold effects emerge at an AI adoption index of 0.43-0.45, where economic impacts increase threefold, and the working-age population's growth effect varies from 0.072 below the threshold to 0.411 above the threshold in the highest industrial upgrading quartile.

Multimodal data fusion for precision customer marketing based on deep learning: service quality perception and loyalty prediction

Xiaojing Nie, Fauziah Sh. Ahmad — Wed, 25 Jun 2025 00:00:00 +0000

Contemporary marketing faces challenges in analyzing complex, multidimensional customer-brand relationships from unprecedented volumes of multimodal data. Traditional analytical approaches inadequately capture this complexity, limiting precision marketing effectiveness. This research develops and validates a comprehensive multimodal data fusion framework utilizing deep learning architectures to enhance service quality perception analysis and customer loyalty prediction. The methodology integrates four data modalities—textual reviews, behavioral patterns, transactional records, and visual content—through specialized neural encoders: CNN for structured data, BERT transformers for textual analysis, LSTM networks for sequential behaviors, and transformer-based encoders for service indicators. Multi-head attention mechanisms and cross-modal feature weighting strategies unify these components while maintaining interpretability through SHAP-based analysis. Experimental validation across 15,420 customers demonstrates substantial performance improvements: service quality prediction (R² = 0.891, MAE = 0.142), customer loyalty classification (F1-score = 0.875, AUC-ROC = 0.923), and churn risk assessment (F1-score = 0.864, AUC-ROC = 0.917), significantly outperforming traditional baselines. Marketing optimization results demonstrate remarkable enhancements: conversion rates (+43.5%), ROI (+56.8%), click-through rates (+81.3%), and revenue per user (+71.1%), all of which are statistically significant (p < 0.001). Customer segmentation analysis reveals that value customers prioritize operational excellence and technical expertise, while regular customers emphasize interpersonal service dimensions. This framework advances multimodal learning theory in marketing contexts, providing practical foundations for next-generation customer relationship management systems. It enables enhanced customer engagement and business value creation through integrated data strategies.

Occupants’ satisfaction with STPV window design in private and open spaces by VR images

Zhan Chen, Nangkula Utaberta, Nadzirah Zainordin — Mon, 14 Jul 2025 00:00:00 +0000

Semi-transparent photovoltaic (STPV) systems have gained increasing attention for their ability to generate electricity while reducing energy consumption compared to conventional windows, addressing climate and energy challenges. However, STPV systems inherently reduce window transparency, which may compromise occupant visual comfort and satisfaction. This study experimentally investigates occupant satisfaction with crystalline silicon (c-Si) STPV windows at different cell coverage ratios (CCR) in private offices and open spaces using virtual reality (VR) technology validated by the Igroup Presence Questionnaire (IPQ). Forty-five participants evaluated six CCR configurations (0%-50%) across two spatial types. Results show VR environments achieved satisfactory presence levels (IPQ: 70.37% private, 70.06% open), validating the methodology. Occupant satisfaction decreased with increasing CCR in both spaces, from 5.11 to 3.00 (private) and 5.89 to 3.22 (open). Open spaces showed significantly higher satisfaction than private offices for 10%-40% CCR, with convergence at 50% CCR. These findings provide design guidance for optimizing STPV integration while maintaining occupant comfort.

Multi-agent reinforcement learning for Bai ethnic traditional dwelling protection in Dali: cultural identity-oriented community relationship optimization and urbanization adaptation algorithm

Xiaohua Qian, Neilson Ilan Mersat, Haslina Hashim, Bemen Wong Win Keong — Sat, 19 Jul 2025 00:00:00 +0000

The preservation of residential architecture from traditional ethnic groups has never faced the types of challenges it does today due to urbanization. These challenges include the insufficient retention of landmarks due to competing stakeholder interests, which often leads to irreversible loss of cultural heritage. This research proposes a new culturally identity-oriented multi-agent reinforcement learning system for the protection of Bai ethnic traditional dwellings in Dali, Yunnan Province. The research combines diverse multi-source data collection approaches, including the building’s architecture and culture, urbanization statistics, and stakeholder networks, and develops an advanced computational framework in which every stakeholder category is embedded as an independent intelligent agent with specific behavioral patterns and autonomous decision-making skills. Specialized deep Q-networks of enhanced Q-value methods that consider cultural identity loss in Q-value calculus through loss function adjustments aimed at balancing cultural preservation and stakeholder appeasement were employed within the framework. Implementation results show performance with an overall accuracy of 89.3% for implementation and 87.2% for cultural preservation effectiveness. Conventional approaches previously achieved significantly lower accuracy within these parameters, 15-25 percentage points. Enhancements in cultural identity increase from a baseline of 58.3% to optimized values of 91.2%, while community satisfaction improves from 54.7% to 86.4%. The framework maintains coordination indices above 85% for all stakeholder groups, showing scalability with over 85% replication success rates for populations between 5,000 and 50,000 residents. This demonstrates theoretical and practical value in the use of AI concerning culturally aware heritage preservation.

AI-based tourist behavior analysis and cultural communication optimization strategies for Shanxi great wall heritage site

Xuehe Hou, Zulhilmi B Paidi — Thu, 10 Jul 2025 00:00:00 +0000

This study analyses tourist behavior and cultural communication optimization strategies of the Shanxi Great Wall heritage site using more sophisticated artificial intelligence technologies. The gaps in heritage tourism are approached by applying machine learning, natural language processing, and multi-objective optimization to exhibit technological management while maintaining cultural integrity. Using a combination of qualitative and quantitative methods, this research gathered data from 1,200 tourists through surveys, interviews, and digital behavior observation as well as social media and online review analysis. Machine learning clustering analysis categorised tourists into five behavioral groups: Heritage Enthusiasts (28.7%), Cultural Explorers (23.4%), Adventure Seekers (19.8%), Quick Visitors (16.2%), and Social Influencers (11.9%). Each segment exhibited distinct engagement patterns and communication preferences. Random Forest outperformed in predicting satisfaction, achieving 87.3% accuracy, followed by Support Vector Machine (84.1%) and Neural Networks (82.6%). AI content optimization’s projected user engagement rate was 43.7% and cultural knowledge transfer effectiveness was improved by 52.1%. The rationalising optimization framework showed marked improvements on various business metrics such as an increase of 47.3% in satisfaction scores, 38.9% in cultural understanding, and a reduction of 29.6% in response times. Validation through pilot implementations proved the framework’s success in integrating conflicting goals of maximising visitor satisfaction, operational efficiency, and preserving cultural elements. This research adds to the growing literature on AI-powered management of heritage tourism and offers actionable recommendations for responsible cultural engagement at heritage sites around the world.

Autonomous mobile robotics in smart warehousing: a cyber-physical systems approach to inventory management

Yaqing Zhang, Julie U. Abellera — Sat, 02 Aug 2025 00:00:00 +0000

Traditional warehouse management systems face unprecedented challenges in the Industry 4.0 era, including escalating e-commerce demands, acute labor shortages, and critical requirements for real-time inventory visibility. Existing solutions fail to deliver the flexibility, scalability, and operational efficiency essential for contemporary supply chain operations. A novel integration framework combining Autonomous Mobile Robots (AMR) with Cyber-Physical Systems (CPS) is presented to enable intelligent, adaptive inventory management in smart warehouse environments. A multi-layered CPS architecture incorporating AMR fleet coordination, real-time data analytics, and digital twin synchronization is proposed. The framework employs distributed task allocation algorithms, dynamic path planning strategies, and predictive inventory optimization models. Implementation leverages edge computing for real-time decision-making and cloud infrastructure for comprehensive data analysis and storage. Experimental validation in industrial environments demonstrates significant performance improvements: 42% enhancement in order fulfillment speed, 35% reduction in inventory holding costs, and 89% accuracy in real-time stock tracking. The system maintained 99.2% uptime reliability while successfully managing 3× peak demand variations. The research advances smart logistics by establishing a scalable, generalizable CPS-AMR framework applicable across diverse warehouse environments. The findings provide actionable guidelines for Industry 4.0 transformation initiatives and establish theoretical foundations for next-generation autonomous warehouse systems.

Digital media-enhanced cultural brand development: creative strategies for urban identity construction

Wei Zhang, Mastika b. Lamat — Mon, 04 Aug 2025 00:00:00 +0000

This study investigates digital media-enhanced cultural brand development through creative strategies for urban identity construction, aiming to develop a comprehensive framework that extends integrated marketing communication theory while preserving cultural authenticity and enabling community participation. A mixed-methods approach incorporated stakeholder surveys (n=420) across four participant groups, systematic analysis of 800 social media posts, examination of 35 institutional platforms, and comparative analysis of 12 international cities, utilizing an AI-enhanced analytical pipeline with natural language processing, computer vision, and machine learning algorithms. Five core creative strategy elements were empirically validated: digital storytelling integration, community-centered content creation, heritage-digital synthesis, interactive cultural experiences, and dynamic community feedback systems. Community-centered content creation achieved the highest community acceptance correlation (r=0.91) and cultural authenticity correlation (r=0.84), while heritage-digital synthesis demonstrated the strongest urban identity impact correlation (r=0.88). Strategic digital media integration significantly amplifies urban cultural brand transmission effectiveness while preserving authentic cultural identity through community-centered approaches, providing urban administrators with evidence-based implementation frameworks for sustainable urban identity construction.

Intelligent optimization of double-helix oil cooling system for outer rotor in-wheel motors based on multi-physics coupling simulation

Bin Xu, Aldrin D. Calderon — Tue, 05 Aug 2025 00:00:00 +0000

Outer rotor in-wheel motors face critical thermal management challenges due to constrained heat dissipation within wheel hubs, limiting their application in electric vehicles. This study addresses the research gap of inadequate cooling solutions for high-power-density motors by developing an innovative double-helix oil cooling system through multi-physics coupling optimization. The proposed framework integrates MotorCAD-Maxwell-Ansys platforms to simultaneously analyze electromagnetic losses, thermal conduction, and fluid dynamics. Key findings demonstrate that the optimized double-helix configuration achieves 28% heat dissipation efficiency enhancement, 17% temperature uniformity improvement, and 5°C peak temperature reduction compared to conventional single-channel systems, while maintaining an acceptable 15% pressure drop penalty. Experimental validation confirms 96.9% correlation with simulation results. This research provides practical thermal management solutions crucial for advancing electric vehicle motor technology.

POA-MLSP: a multi-dimensional learning analytics framework for predicting CET4 writing performance based on a production-oriented approach and student engagement patterns

Yu Li, Nur Ainil BT. Sulaiman, Halizah BT. Omar — Wed, 06 Aug 2025 00:00:00 +0000

Contemporary College English Test Band 4 (CET-4) writing instruction faces significant challenges in accurately predicting student performance and providing timely pedagogical interventions. This study develops and validates the Production-Oriented Approach Multi-Dimensional Learning Analytics Framework for Student Performance (POA-MLSP) for predicting CET-4 writing performance across five dimensions through systematic integration of Production-Oriented Approach (POA) theory and Self-Determination Theory (SDT)-based engagement modeling. The framework implements a four-layer architecture incorporating Feature Adaptive Selection Mechanism and SDT-Based Engagement Dynamic Modeling algorithms. Validation involves 124 students during a 16-week semester, collecting multi-source data including Jacobs' five-dimensional assessments, Utrecht Work Engagement Scale-Student (UWES-S) engagement measurements, classroom observations, and digital platform interactions across experimental and control groups. POA-MLSP achieves R² = 0.75 overall prediction accuracy, outperforming linear regression (R² = 0.58), random forest (R² = 0.66), and support vector machines (R² = 0.63) by 17-29%. Content prediction reaches highest accuracy (R² = 0.78), while the framework identifies five distinct engagement profiles and achieves 78.4% ± 2.1% early warning accuracy with 79.8% ± 2.9% teacher satisfaction. Educational theory-guided algorithms significantly enhance prediction performance while maintaining pedagogical interpretability, enabling proactive intervention through early warning systems with minimal implementation burden for authentic educational applications.

Digital-enhanced talent cultivation mechanisms in entrepreneurial universities: an AI-integrated multi-level analysis of student entrepreneurial intentions

Zhiyuan Lyu, Yusri Kamin — Thu, 07 Aug 2025 00:00:00 +0000

This study examines talent cultivation in entrepreneurial universities and investigates how formal and informal factors affect students' entrepreneurial intentions. Analysis of 782 students from eight Chinese universities, enhanced by machine learning predictive models, reveals that informal culture, particularly entrepreneurial culture ( B=0.36), combined with AI-powered personalized learning pathways (B=0.28), correlates significantly with entrepreneurial intentions. The interaction between curriculum and culture (B=0.23) suggests that educational efforts achieve greater effectiveness within supportive cultural environments. This research contributes to entrepreneurial talent development through institutional theory lenses and offers a contextual framework for universities to strategically shape entrepreneurial attitudes amid rapid changes in Chinese higher education.

Research on an intelligent decision support system for enterprise organizational change in the digital economy environment

Kexin Zhang — Fri, 08 Aug 2025 00:00:00 +0000

This investigation outlines a new intelligent system to assist in decision-making for enterprise organisational changes in the context of the digital economy. The innovations of this study are threefold: First, the creation of a multi-dimensional decision model defined by the real-time indicators from the digital economy, as well as traditional metrics of organisational change for structural evolution. Second, the application of a hybrid intelligent algorithm that incorporates deep learning with knowledge graphs enables the processing of both structured and unstructured data at the enterprise level, thereby offering broader decision-making support than standard systems. Third, the development of a system that provides optimised decision recommendations based on what happens after the decision is implemented, thus closing the gap between system design and reality. Results from practical tests conducted in several enterprises substantiate that the proposed system has 35% greater efficiency in making decisions and 42% lower risks in implementing organisational changes than the traditional methods. This development has a considerable impact on the teaching and practice of intelligent decision support in enterprise digital transformation, posing a new approach to managing organisational changes in the digital economy.

FNet-GPT: Fourier-based lightweight transformer for emotion-aware text generation using GPT

Atul Kachare, Chandrashekhar Goswami, Ashutosh Gupta, D.S. Chouhan — Wed, 13 Aug 2025 00:00:00 +0000

Neural story generation models have two significant challenges: (1) coherence over narrative structure, especially long-range dependencies, and (2) emotional coherence and consistency, generally producing redundant or incoherent narration. A new, emotionally intelligent two-stage short story generation model is presented by combining GPT-2 with a tailored FNET model, a light transformer architecture substituting standard self-attention with Fourier Transform layers to improve semantic and emotional relationship capture in text. The first stage employs GPT-2 to generate a list of input candidate sentences, a question, an answer, and an emotional state. The candidate sentences are then filtered using an emotion classifier from DistilRoBERTa to keep only those that adhere to a desired emotional tone. The filtered sentences are then fed into a fine-tuned FNET model, which examines inter-sentence relationships and enforces emotional coherence to generate a coherent and emotionally engaging narrative. An empirical comparison using three benchmark datasets demonstrates the system's superiority over earlier state-of-the-art approaches. The FNET model achieves 0.3093 in BLEU-1, outperforming Plan-and-Write (0.0953) and T-CVAE (0.2574), with an enhanced narrative quality and lexical coherence with human-written narratives. The story coherence and emotion retention accuracies are 85%, 67%, and 60% for Visual7W, ROCStories, and Cornell Movie Dialogs datasets.

AI-driven marketing innovation in educational technology: a multi-dimensional analysis of virtual sales personnel and intelligent promotion strategies on user acceptance and engagement

Chuntie Chen, Nor Hidayati Binti Zakaria, Wei Deng, Xiaoli Xu, Youyu Xu — Thu, 14 Aug 2025 00:00:00 +0000

This study investigates the impact of AI-driven marketing innovations on user acceptance and engagement in educational technology contexts, examining how virtual sales personnel characteristics and intelligent promotion strategies influence behavioral outcomes through psychological mechanisms. An explanatory sequential mixed-methods design was employed, combining structural equation modeling analysis of survey data from 650 educational technology users with thematic analysis of 45 semi-structured interviews. Machine learning algorithms, particularly XGBoost (AUC=0.89), were utilized to predict user acceptance patterns and identify five distinct user segments. Trust emerged as the critical mediating mechanism between AI anthropomorphism and user acceptance, accounting for 76.5% of the total effect. Personalization capabilities demonstrated the strongest impact on continuous engagement (β=0.52, p<0.001). Qualitative analysis revealed three overarching themes: intelligent companion experience (82.2% prevalence), personalization value perception (88.9%), and privacy-convenience trade-offs (68.9%). The validated framework provides educational technology enterprises with actionable guidelines for implementing AI marketing systems that balance technological sophistication with humanization principles through moderate anthropomorphism and progressive personalization strategies. This research extends the Technology Acceptance Model by integrating AI-specific constructs, including algorithm trust and perceived intelligence, offering novel theoretical insights and empirical evidence for optimizing human-AI interactions in educational marketing contexts. AI fundamentally transforms educational technology marketing through trust-based mechanisms, requiring careful balance between innovation and humanization for sustainable adoption.

Research on a virtual teacher personalized interaction model integrating affective computing and multi-agent systems

Rili Dang, Noorazman Abd Samad — Mon, 18 Aug 2025 00:00:00 +0000

This research develops a novel virtual teacher personalized interaction model integrating multimodal affective computing with multi-agent coordination mechanisms to address fundamental limitations in emotional intelligence and adaptive capabilities within contemporary educational technology systems. A three-layer distributed architecture was implemented, incorporating synchronized multimodal emotion recognition through confidence-weighted fusion of facial, vocal, and textual data streams, Byzantine Fault Tolerant consensus algorithms for coordinated multi-agent decision-making, and dynamic personality adaptation mechanisms based on Big Five psychological modeling. Experimental validation employed 500 participants across diverse educational contexts using established emotion recognition benchmarks supplemented with domain-specific educational interaction datasets. The multimodal emotion fusion component achieved 91.2% recognition accuracy, with overall system performance reaching 89.7% under realistic educational conditions while demonstrating substantial educational effectiveness improvements, including 43% higher learner engagement scores, 37% emotional satisfaction enhancement, 30% learning effectiveness increase, and 40% knowledge retention improvement compared to traditional virtual teaching approaches. Multi-agent coordination exhibited superior decision quality with 31% improvement over single-agent baselines, though personality adaptation effectiveness varied significantly across learner populations with 88% success rates for extraverted individuals compared to 65% for high-neuroticism learners. The integrated approach successfully bridges the emotional intelligence gap in virtual educational systems through sophisticated technological convergence, establishing theoretical foundations for distributed educational intelligence while revealing important implementation challenges. This research enables the development of emotionally responsive virtual teachers capable of sustained personalized instruction across diverse educational contexts, though deployment requires careful consideration of privacy protection and institutional adaptation requirements for broader educational technology transformation.

Optimizing blended learning through AI-powered analytics in digital education platforms: an empirical framework

Fengrui Zhang, I Wayan Subagia, Luh Putu Artini, Dessy Seri Wahyuni — Tue, 19 Aug 2025 00:00:00 +0000

This study proposes an empirical framework for enhancing blended learning through Artificial Intelligence (AI)-powered analytics in digital education platforms. The research employs a mixed-methods approach, examining 250 undergraduate business students engaged in blended learning courses over one semester. Quantitative data from platform analytics, academic performance metrics, and structured questionnaires are analyzed using descriptive statistics, regression analysis, and machine learning algorithms. Results demonstrate significant improvements in learning outcomes, with overall academic performance increasing from 72.4% to 81.7% (p < 0.001). Critical thinking skills improve by 24.3%, collaborative abilities by 31.2%, and digital literacy by 28.7%. Cluster analysis reveals three distinct learner profiles, with engagement patterns serving as strong predictors of academic success (R² = 0.584). AI-powered predictive models achieve 83.7% accuracy in identifying at-risk students by week four, enabling targeted interventions that improve outcomes by 67%. Platform engagement frequency emerges as the strongest predictor (β = 0.42, p < 0.001). Critical engagement periods occur during weeks 3-5 and 10-12. The framework integrates multiple learning theories within AI-enhanced contexts and provides practical guidance for platform optimization, instructional design, and policy development. Findings emphasize that successful blended learning requires purposeful technology integration with pedagogical principles, continuous engagement monitoring, and personalized support mechanisms.

AI-enhanced strategic management of foreign direct investment in China: decoding the mediating effects of labor productivity and infrastructure development on economic growth

Zhibin Jia, Oyyappan Duraipandi — Fri, 22 Aug 2025 00:00:00 +0000

The integration of artificial intelligence into foreign direct investment management represents a paradigm shift in strategic decision-making, particularly as traditional analytical frameworks prove inadequate for navigating complex economic environments characterized by volatility, uncertainty, complexity, and ambiguity. This study develops an AI-enhanced strategic management framework for optimizing foreign direct investment (FDI) allocation decisions while examining the mediating roles of labor productivity and infrastructure development in facilitating economic growth outcomes in China's digital transformation context. Employing a mixed-methods approach combining panel data analysis from 30 Chinese provinces (2010-2023) with machine learning algorithms including Random Forest, Support Vector Machines, and Neural Networks, the research integrates traditional econometric techniques with AI-powered predictive modeling to capture complex non-linear relationships. AI-enhanced management achieves 21% higher prediction accuracy (R²=0.743) compared to traditional methods (R²=0.614) while reducing processing time by 78%. Labor productivity mediates 41.2% and infrastructure development 21.0% of the total effect on economic growth, with significant synergistic interactions (β=0.087, p<0.01) amplifying individual contributions. The findings establish AI integration as essential for modern FDI management, providing theoretical advancement through the four-layer architectural framework and practical implementation guidelines. The research demonstrates that successful AI-enhanced FDI strategies require simultaneous optimization of technological capabilities, human capital development, and infrastructure readiness.

Research on intelligent regulation mechanisms of learner cognitive load in digital learning environments

Jianxu Zhai, I Gusti Putu Sudiarta, Made Hery Santosa, I Wayan Puja Astawa — Sat, 23 Aug 2025 00:00:00 +0000

This research develops an intelligent cognitive load regulation framework for digital learning environments in the context of educational policy reforms. After China's Double Reduction Policy took effect, tutorial-concentrated schooling evolved into technology-facilitated learning, putting unimaginable cognitive burdens on students. In response, the research combines cognitive load theory with adaptive technologies to resolve these issues through real-time recognition of cognitive states and personalized interventions. Based on the mixed-methods design with 320 Dongcheng District students, the research uses established measures such as NASA-TLX adapted to e-learning environments to assess multidimensional patterns of cognitive load. The smart regulation system shows significant efficacy with lower socioeconomic students posting 15.3-point improvements in academic scores, task accomplishment rates enhanced by 32%, and the level of cognitive loads decreased by 23.1% on average across various types of learners. The system can recognize with 87.3% accuracy and respond in 234 milliseconds, thus facilitating timely interventions. Self-paced review activities yield 91.2% success rates, while collaborative tasks remain problematic at 68.4% success rates. The results extend cognitive load theory with dynamic adaptation capacities needed for self-managed digital learning. The present study provides evidence-based practice to maximize cognitive experiences of e-learning, facilitating education equity objectives while developing core self-regulated learning skills in post-reform education systems.

Toward sustainable power with floating solar at Near East University Lake, Northern Cyprus

Youssef Kassem, Hüseyin Çamur, MohamedAlmojtba Hamid Ali Abdalla — Mon, 25 Aug 2025 00:00:00 +0000

Floating solar photovoltaic (FPV) systems have become a desirable research topic for optimization and development. The primary objective of the current study is to optimize an FPV at Near East University Lake in Northern Cyprus, aiming to enhance energy production and mitigate negative environmental impacts. Besides, the potential for energy generation and economic feasibility of various design configurations related to fixed and tracked PV systems and coverage area (45, 60, 75, and 90%) were investigated. The results demonstrated that the increase in coverage area indeed increased energy yield due to the increase in the number of panels. The 90% coverage area, for instance, reduces the cost of energy production to 0.0176 USD/kWh and produces a very respectable increase in energy yield. According to the techno-economic analysis, the reduction of GHG emissions can range from 330 to 659 tCO2/year, depending on the coverage area. The value of NPV demonstrates the system's long-term sustainability and profitability, while the basic payback period remains relatively consistent across all coverage percentages, ranging from 3.19 to 3.20 years. Thus, this research provides valuable insights into how floating solar technology can be integrated with water conservation and sustainable energy production, which can greatly aid in achieving renewable energy targets and reducing water evaporation losses.

Personalized learning pathways in AI-powered dubbing applications for speaking proficiency enhancement: a systematic review

Ruilin Zhao, Hanita Hanim Ismail, Ahmad Zamri Mansor — Thu, 28 Aug 2025 00:00:00 +0000

The integration of artificial intelligence in language education has revolutionized pedagogical approaches, with AI-powered dubbing applications emerging as promising tools for developing speaking proficiency through personalized learning pathways. This systematic review synthesized evidence from 38 empirical studies involving 4,327 participants to evaluate the effectiveness of personalized learning pathways within AI-powered dubbing applications for Business English speaking proficiency enhancement. Following PRISMA guidelines, comprehensive searches across seven databases identified peer-reviewed studies published between 2019-2024, with quality assessment employing Cochrane risk-of-bias tools and meta-analysis conducted where appropriate. The analysis revealed substantial improvements in pronunciation accuracy (Cohen's d=1.82, 95% CI: 1.65-1.99) and fluency development (d=1.46, 95% CI: 1.29-1.63), with intermediate-level learners demonstrating 68.4% greater gains compared to advanced learners. Subgroup meta-analysis confirmed neural network superiority over collaborative filtering approaches, achieving 87.3% accuracy in pronunciation feedback. Publication bias assessment revealed asymmetrical distribution (p=0.031), though trim-and-fill analysis indicated minimal impact on primary conclusions. Cost-effectiveness analyses demonstrated significant advantages, requiring $15-25 per student annually compared to $180-240 for equivalent individual tutoring. Cultural engagement patterns aligned with Hofstede's dimensions theory, where East Asian learners showed higher completion rates but lower self-efficacy scores. Despite documented learning plateau effects after 4-6 weeks, AI-powered dubbing applications demonstrate significant potential for enhancing speaking proficiency, though optimal implementation requires hybrid approaches integrating human pedagogical expertise with technological affordances to address cultural contextualization and sustained engagement challenges.

Research on intelligent optimization mechanisms of financial process modules through Machine Learning-enhanced collaborative systems in digital finance platforms

Ting Wang, Grace R. Tobias — Tue, 02 Sep 2025 00:00:00 +0000

This research addresses the critical need for intelligent optimization mechanisms in financial process modules by developing a machine learning-enhanced collaborative system designed for digital finance platforms, aiming to bridge theoretical advances in human-machine collaboration with practical applications in financial process optimization. A sophisticated multi-layered architecture integrating machine learning capabilities with human decision-making processes was developed, incorporating advanced ensemble algorithms, multi-objective optimization techniques, and adaptive learning mechanisms. The system was validated across three real-world scenarios. These included credit risk assessment using 2.26 million Lending Club records, anti-money laundering with 6.3 million FinCEN transactions, and customer service optimization with 1.8 million banking interactions. The collaborative system achieved significant improvements. Cost reduced by 28.4% and accuracy increased by 15.3% in credit risk assessment. AML efficiency improved by 256%, and AUC-ROC increased from 0.847 to 0.923. Processing time was reduced from 4.2 days to 1.8 days while maintaining regulatory compliance, resulting in a 44.8% return on investment in the first operational year. The learning collaborative approach efficiently combines human knowledge and AI, outperforming regular computerized methods as well as purely human strategies and maintaining long-term system improvement through its adaptive learning capability. This study provides practical toolkits for financial institutions to further explore AI in process optimization, aiming to achieve sustainable competitive advantages and compliance, while also ensuring operational efficiencies.

Intelligent collaboration and artistic co-creation: a study on the enhancement mechanism of social well-being through AI-enabled intergenerational integration

Wanyi He, Anuar Bin Ahmad, Nasruddin Yunos, Bingbing Chen — Wed, 03 Sep 2025 00:00:00 +0000

This study investigates how AI-enabled intergenerational artistic co-creation enhances Social well-being through a mixed-methods approach involving 120 participants across younger (15-25) and older (65+) age cohorts. The findings reveal a novel "triangulated collaboration model" wherein AI functions as both creative catalyst and communicative bridge between generations. Empirical results demonstrate statistically significant improvements: technological engagement convergence increased from 62% to 79% among older adults (p < .001), bidirectional knowledge transfer showed 28.7-point gains in cultural knowledge and 32.5-point gains in technical proficiency, and creative innovation scores improved by 47.2% in intergenerational groups compared to 22.9-28.6% in age-homogeneous groups. We identify multilevel enrichment mechanisms: at the individual (psychological well-being, self-efficacy, creativity), relational (communication, empathy, social capital), and community (inclusive behavior, community participation, cultural heritage preservation) levels. The Intelligent Collaborative Enhancement Model (ICEM) is a theoretical model that outlines how technological adaptability, creative co-construction, and mutual learning form "generative integration spaces." Policy implications from this research are for educational, cultural, and social welfare policies, considering how the utilization of technological mediation can foster strong intergenerational relationships within a more age-diverse society.

Research on risk control and sustainability strategies of AI-driven big data analytics in LEAN manufacturing equipment R&D

ChengHsien Tsai, Oyyappan Duraipandi, Dhakir Abbas Ali — Fri, 05 Sep 2025 00:00:00 +0000

The convergence of artificial intelligence (AI) and LEAN manufacturing principles presents unprecedented opportunities for operational excellence while introducing complex risk management and sustainability challenges. Addressing the critical research gap in quantitative AI-LEAN integration models. This research develops an integrated framework for implementing AI-driven big data analytics in LEAN manufacturing equipment R&D, addressing the critical gap between technological capabilities and sustainable manufacturing practices. We used three research methods: theoretical modelling, empirical validation with the SECOM semiconductor dataset, and 12-month field testing across three manufacturing facilities. This mixed-methods approach quantifies the synergistic effects of AI-LEAN integration. The framework incorporates hierarchical risk taxonomy, real-time anomaly detection algorithms achieving 93.5% accuracy, and multidimensional sustainability metrics. Results demonstrate substantial improvements: 36.1% increase in overall equipment effectiveness, 58.9% reduction in setup times, and 31.4% decrease in carbon footprint, energy intensity reduced by 30%, employee safety incidents decreased by 67%, and job satisfaction increased by 15%, achieving synergistic optimization of environmental benefits and social value. Risk prediction models achieved 91-96% accuracy across different categories, while maintaining sub-50ms inference times for real-time applications. The AI-enhanced system outperformed traditional LEAN implementations by 1.81x in continuous improvement rates and achieved payback in 13 months versus 23 months for conventional approaches. Financial analysis reveals 319.4% ROI over five years, validating the economic viability alongside environmental benefits. This research establishes a replicable paradigm for sustainable smart manufacturing, demonstrating that advanced analytics can simultaneously enhance operational efficiency, risk management, and environmental stewardship while preserving LEAN's human-centric values.

Integrated scheduling of jobs, tools, and AGVs in FMS with non-identical machines using a recurrent neural network

Swapnil Janardan More, Naveen Kumar — Wed, 10 Sep 2025 00:00:00 +0000

In a flexible manufacturing system (FMS), scheduling jobs and tools across non-identical machines, integrating automated guided vehicles (AGVs), and considering multi-objective functions, constitutes a significant obstacle for typical mathematical optimization techniques. Herein, we consider scheduling jobs, tools, and AGVs in an FMS that consists of three non-identical machines. The multi-objective functions targeted are tooling cost minimization and makespan reduction. The non-identical machines' processing rates are specified in the ratio of 1:1.2:1.4. Each of the tools (T1, T2, and T3) is available in a single mode, with T1 being more expensive than T2, which is more expensive than T3. To address such a complex optimization problem, we use a Recurrent Neural Network (RNN) and an Improved version to obtain near-optimum solutions and evaluate such algorithms' comparative performance. The average computation time to determine the optimal sequence was reduced from 10.33 minutes to 6.24 minutes (for a 4-job problem) as we employed the Improved RNN algorithm instead of the RNN algorithm.

Deep Learning-based anomaly detection in stock markets and business decision support

Changjiang Dai — Mon, 15 Sep 2025 00:00:00 +0000

The increasing complexity and volatility of modern financial markets necessitate advanced anomaly detection systems that can identify irregular patterns, which may signal market manipulation, systemic risks, or emerging crises. This research presents a comprehensive deep learning framework for real-time anomaly detection in stock markets, integrated with business decision support systems to enhance risk management and regulatory compliance. We propose and evaluate four distinct deep learning architectures: LSTM-Autoencoder, Variational Autoencoder (VAE), Transformer-based models, and an ensemble approach, utilizing high-frequency trading data from major stock exchanges spanning 2019-2024. Our methodology incorporates multi-dimensional feature engineering, including technical indicators, market microstructure variables, and sentiment analysis, processed through advanced normalization techniques. The experimental results demonstrate that the Transformer-based ensemble model achieves superior performance with an F1-score of 0.89 and AUC of 0.94, representing a 43.5% improvement over traditional methods (F1=0.62 for ARIMA-GARCH) and 17% improvement over standalone machine learning approaches (F1=0.76 for XGBoost). The system successfully detected 92% of major market anomalies with a 15-minute average early warning time while maintaining a false positive rate below 3%. Furthermore, the integration with decision support systems yielded a 34% improvement in risk-adjusted returns for test portfolios, reducing decision-making time by 67.3% (from 98s to 32s) and achieving cost savings of $35.2M monthly across deployed institutions. This research contributes to financial technology by bridging the gap between advanced deep learning techniques and practical business applications, offering a scalable solution for market surveillance and risk management in increasingly complex financial ecosystems.

Innovative applications of big data and simulation technologies in the optimization design of crash safety for autonomous vehicles: a systematic review from a biomechanical aspect

Lingxiao Sun — Fri, 19 Sep 2025 00:00:00 +0000

The rapid development of autonomous vehicles (AVs) has intensified the demand for advanced strategies to guarantee crash safety in increasingly complex traffic environments. Traditional design methods, reliant on physical crash tests and limited empirical data, are insufficient to capture the full spectrum of biomechanical responses during collisions. This systematic review synthesizes recent advances in the integration of big data analytics and simulation technologies for optimizing collision safety, with a particular focus on biomechanical modeling. Big data enables the large-scale collection and analysis of heterogeneous data sources—including vehicle sensors, physiological signals, and traffic dynamics—supporting the construction of high-fidelity injury prediction models. Simulation methods, such as finite element analysis (FEA), multi-body dynamics (MBD), and parametric optimization, facilitate precise evaluation of occupant kinematics, stress distributions, and tissue-level injury mechanisms. Furthermore, emerging applications of machine learning, digital twin systems, and biomimetic design demonstrate substantial potential for improving active and passive safety. This review highlights the synergistic role of biomechanics, data science, and simulation technologies in shaping the next generation of collision protection systems. Finally, it identifies key challenges—including data privacy, model accuracy, and computational efficiency—and proposes future directions toward multi-scale biomechanical modeling, AI-driven optimization, and cross-disciplinary integration for safer and more adaptive autonomous driving systems.