Predictive modeling of transverse cracking in continuously reinforced concrete pavement: a machine learning approach

Engineering Research Express, Journal Year: 2025, Volume and Issue: 7(1), P. 015106 - 015106

Published: Jan. 6, 2025

Abstract Accurate prediction of transverse cracking in Continuously Reinforced Concrete Pavement (CRCP) is critical for improving infrastructure management procedures and preserving the road network’s long-term durability safety. This paper conducts a thorough analysis into predicting CRCP using machine learning approaches. The research involved meticulous data preparation, feature selection, evaluation various models to identify most effective predictor. Key variables such as pavement age, total thickness, temperature, freeze index, traffic volume, precipitation, initial International Roughness Index (IRI) were analyzed their impact on occurrences. Sensitivity was conducted assess influence individual input model predictions. Results indicated that cubic Support Vector Machine (SVM) outperformed other models, demonstrating exceptional predictive accuracy. Furthermore, sensitivity revealed significant correlations between occurrences, emphasizing importance considering holistic range factors engineering maintenance strategies. Our findings, which provide insights intricate interactions distress, help create tailored treatments methods, optimized crack sealing schedules, improved reinforcement strategies, use high-performance materials, minimizing enhancing performance.

Language: Английский

---

Understanding Older Adults’ Intention to Adopt Digital Leisure Services: The Role of Psychosocial Factors and AI-Based Prediction Models

Healthcare, Journal Year: 2025, Volume and Issue: 13(7), P. 785 - 785

Published: April 1, 2025

Background/Objective: As the global aging population grows, digital leisure services have emerged as a potential solution to improve older adults' social engagement, cognitive stimulation, and overall well-being. However, their adoption remains limited because of literacy gaps, psychological barriers, varying levels adaptability. This study aims analyze predict intention adopt by integrating psychosocial factors, demographic characteristics, adaptability using artificial intelligence (AI)-based predictive models. Methods: utilized data from 2022 Urban Policy Indicator Survey conducted in Seoul, South Korea, selecting 2239 individuals aged 50 years above. A two-step clustering approach was employed: hierarchical estimated optimal number clusters, K-means finalized segmentation. An neural network (ANN) model applied likelihood incorporating variables. Logistic regression used for validation, performance assessed through accuracy, precision, recall, F1-score. Results: Four distinct clusters were identified based on media engagement. Cluster 3 (highly educated males 60s with family support) showed highest probability (84.35%) despite low 4 (older women high usage) exhibited lower structured services. The ANN achieved an classification accuracy 85.2%, highlighting key determinant adoption. Conclusions: These findings underscore need targeted policy interventions, including tailored education programs, intergenerational training, simplified platform designs enhance accessibility. Future research should further explore factors influencing validate AI-based predictions real-world behavioral data.

Language: Английский

---

A comprehensive analysis of stroke risk factors and development of a predictive model using machine learning approaches

Molecular Genetics and Genomics, Journal Year: 2025, Volume and Issue: 300(1)

Published: Jan. 24, 2025

Stroke is a leading cause of death and disability globally, particularly in China. Identifying risk factors for stroke at an early stage critical to improving patient outcomes reducing the overall disease burden. However, complexity requires advanced approaches accurate prediction. The objective this study identify key develop predictive model using machine learning techniques enhance detection improve clinical decision-making. Data from China Health Retirement Longitudinal Study (2011–2020) were analyzed, classifying participants based on baseline characteristics. We evaluated correlations among 12 chronic diseases applied algorithms stroke-associated parameters. A dose–response relationship between these parameters was assessed restricted cubic splines with Cox proportional hazards models. refined model, incorporating age, sex, factors, developed. patients significantly older (average age 69.03 years) had higher proportion women (53%) compared non-stroke individuals. Additionally, more likely reside rural areas, be unmarried, smoke, suffer various diseases. While correlated (p < 0.05), correlation coefficients generally weak (r 0.5). Machine identified nine associated risk: TyG-WC, WHtR, TyG-BMI, TyG, TMO, CysC, CREA, SBP, HDL-C. Of these, SBP exhibited positive risk. In contrast, TMO HDL-C reduced fully adjusted elevated CysC (HR = 2.606, 95% CI 1.869–3.635), CREA 1.819, 1.240–2.668), 1.008, 1.003–1.012) increased risk, while 0.989, 0.984–0.995) 0.99995, 0.99994–0.99997) protective. nomogram demonstrated superior accuracy, Harrell's C-index individual predictors. This identifies several significant presents that can high-risk Among them, WHtR positively whereas opposite. serves as valuable decision-support resource clinicians, facilitating effective prevention treatment strategies, ultimately outcomes.

Language: Английский

---

Institutional Effectiveness and Economic Development: A Machine Learning Approach With Empirical Modelling

American Journal of Economics and Sociology, Journal Year: 2025, Volume and Issue: unknown

Published: March 11, 2025

ABSTRACT Criminal behaviour and its societal impact remain critical challenges in developing economies, where successful rehabilitation directly influences community well‐being social progress. India's distinct federal structure provides a compelling research setting, as states exercise considerable autonomy programs while operating under unified legal framework. Our methodology combines machine learning with dynamic panel estimation to analyse institutional effectiveness across 29 Indian (2002–2021), examining 14 dimensions of capacity. This comprehensive analysis explores how economic conditions, mechanisms, interact determine reform success. The findings reveal that emerges not from isolated or improvements but through their systematic integration. A pivotal policy shift 2016 demonstrated implemented reforms, strengthening foundations achieved marked outcomes. deepens our understanding capacity, resource allocation, implementation strategies shape success resource‐constrained environments. We identify key determinants by analysing variations identical frameworks. These insights advance knowledge economies can design effective harness growth development enhance

Language: Английский

---