446494 research outputs found
Sort by
Hydrogeochemical characterization of groundwater in the Nekor-Ghiss plain, Morocco
No full textThe coastal Nekor-Ghiss aquifer, located on the northeastern Mediterranean coast of Morocco, constitutes an essential freshwater resource for the city of Al Hoceima, supplying water for agriculture and domestic use across an area of approximately 100 km2. This study provides a comprehensive hydrogeochemical characterization of 25 groundwater samples collected in May 2022, combining physico-chemical analyses with Piper diagram classification to identify dominant water facies and sources of contamination. The results show that the waters are highly mineralized, with electrical conductivity ranging between 1825 and 12720 µS/cm. Three distinct hydrochemical facies were identified: a predominant sodium chloride (NaCl) facies (84%) and a sodium sulfate facies. Marine intrusion and evaporite dissolution particularly affect wells W16 and W17, as evidenced by their high concentrations of Na+ (2572.8 mg/L) and Cl- (3961.8 mg/L). Agricultural contamination was detected in well W21, where nitrate concentrations exceed the WHO limit of 50 mg/L, while well W23 shows critical nitrite pollution (25.56 mg/L), indicating localized contamination. This study highlights the complex interactions between natural geochemical processes and anthropogenic pressures in Mediterranean coastal aquifers and provides essential baseline data for implementing sustainable groundwater management strategies in the region
From Index to Decision: A Critical Review of Water Quality Indices Applied to Moroccan Waters (2018–2025)
No full textWater Quality Indices (WQIs) are widely used to synthesize information on water quality and support decision-making. In Morocco, their application has grown significantly in recent years for the assessment of surface and groundwater. However, the diversity of indices, parameters used, weighting schemes, and reporting practices limits the comparability and robustness of results. This study offers a critical analysis of the applications of WQIs to Moroccan waters over the period 2018–2025, based on a systematic review of 244 articles indexed in Scopus. The results highlight significant methodological heterogeneity, notable classification differences for comparable contexts, and a lack of transparency regarding the management of missing data and uncertainties. In order to strengthen the scientific and decision-making value of WQIs, a minimal harmonization framework adapted to the semi-arid Moroccan context is proposed, based on a common set of parameters, explicit data processing rules, double reading by reference indices, and standardized reporting
Numerical and Analytical Study of Crack Propagation in Pressure Vessels: Linear and Non-Linear Behaviors
No full textThis study investigates the cracking of pressure vessels using advanced numerical methods. It examines the linear and non-linear behavior of materials by studying the evolution of Von Mises stresses, the stress intensity factor and the J-integral. The study emphasizes the integration of numerical models with fracture mechanics, specifically, using the finite element method and the meshless method. The aim is to identify the limitations and complementarities of these approaches to improve the modeling of cracked pressure vessels
Comparative Study of Student Academic Outcomes and Behavioral Patterns through Data-Driven Approaches
No full textPredicting student performance and understanding behavioral patterns have become central themes in modern educational research, particularly with the rise of digital and blended learning environments. The growing availability of data from Learning Management Systems (LMS) and online learning platforms has led to a wide array of data-driven and AI-based approaches for analyzing academic outcomes and learner behaviors. This paper presents a comprehensive comparative review of existing intelligent learning models, examining how various studies utilize behavioral indicators, such as engagement, interaction patterns, and study habits, to predict student performance. By synthesizing findings across diverse methodologies and datasets, the review highlights current trends, strengths, limitations, and research gaps, offering educators and researchers valuable insights for developing more effective, data-informed student support strategies
Changes in EEG recurrence rate during REM sleep caused by Parkinson’s disease
No full textParkinson’s disease is known to be closely associated with sleep and breathing disorders. We examined three groups of subjects: 15 apparently healthy patients, 10 patients with obstructive sleep apnea (OSA), and 25 patients with Parkinson’s disease (PD). For each subject, two full-night polysomnography recordings were recorded at intervals of 1–3 days. Polysomnographic electroencephalography (EEG) recordings were analyzed using quantitative recurrence analysis. The ratio of the average recurrence rate for REM sleep stage to other sleep stages was calculated. The PD group showed statistically significant differences () with both control and OSA groups. However, statistically significant differences were not found between the control and OSA groups. The main conclusion is that Parkinson’s disease alters the fundamental properties of brain activity during REM sleep, leading to an increased probability of returning EEG values to the previous values in time series
A Comprehensive Review of VertexML: A Model Training Platform
No full textThis paper presents a comprehensive survey of automated machine learning (AutoML) techniques, with a focus on meta-learning, hyperparameter optimization, and neural architecture search (NAS). Rather than proposing or evaluating a full AutoML platform, this work synthesizes insights from 49 influential research papers, organizing them into methodological categories and highlighting their contributions to the evolution of ML automation. The survey also analyzes trends across model types, optimization strategies, and publication patterns. Based on this review, the paper identifies research gaps and outlines key directions for future development of unified, scalable, and interpretable AutoML systems. This survey is intended to provide a structured foundation for researchers working toward improved ML automation pipelines
AI-Powered personalization and its impact on sustainable consumer behaviour in digital marketing
No full textWhile AI-driven personalization is revolutionizing e-commerce, it must be matched with sustainability. This paper has explored how AI, utilizing recommendation systems, machine learning and natural language processing, can process information in order to nudge pro-environmental choices and mitigate waste and overconsumption. The focus is on how AI will contribute to greener purchases, less returns and a more circular economy. With mixed methods, the case studies evaluate platforms blending personalization with sustainability across a set of key performance indicators e.g. conversion rates, retention, carbon footprint reduction and product lifecycle optimization. Benefits are evident, but issues of privacy, bias and ethics persist. The research highlights the importance of ethical, sustainable AI that can deliver personalized content without adding to environmental issues
Design and implementation of robot-assisted systems for non-destructive testing
No full textWith wide application of new materials, structures, and technologies in modern industry, testing objects were no longer limited to conventional materials and common shapes. Demands of testing new materials and complex shapes bring challenges to researchers in non-destructive testing area. Combining robots, which have been widely used in industry, with the non-destructive testing technology can replace manual operation and improve testing precision. Additionally, the robot-assisted systems can enhance efficiency and safety of the testing process. The researchers have carried out numerous designs and implementations to combine robots with non-destructive testing devices. This article presents four non-destructive testing systems with robot, including single-arm robot holding a transducer, single-arm robot holding a tested part, and twin-arm robot holding two transducers. In conclusion, application scopes are analyzed to help users select appropriate systems according to sizes, materials and defect types
Handwritten input to speech conversion using transfer learning
No full textSystem developed to transform handwritten text into clear, natural sounding speech attained by using a structured multi-step process. The steps include image preprocessing,clean and enhanced output,elimination of noise and improvement of readability.The unrefined input text is refined using Natural Language Processing(NLP).This refined text is transformed into human-like speech using advanced conversion methods.The system ensures improved recognition,accuracy,effective noise reduction and contextual correction providing smooth and natural speech output.The system is highly useful for education,healthcare,assistance and particularly for aiding visually challenged individuals.The combination of computer vision,deep learning,and language preprocessing bridges the gap between handwritten information and digital accessibility
Characterization and Adsorption Performance of Teak Sawn Derived Activated Carbon for Cr(VI) Removal from Batik Industry Effluents
No full textSawn timber waste requires proper disposal management owing to its potential to pollute water. Teak (Tectona grandis L.) sawn timber residue can be converted to activated carbon for heavy metal adsorption. This study aimed to characterize T. grandis activated carbon, determine its maximum adsorption capacity for Cr(VI) in batik wastewater, and evaluate the adsorption isotherm models. The sawn teak was carbonized, chemically activated with 1 M HCl for 24 h. Characterization was performed using SEM and FTIR. Adsorption experiments were conducted with 0.5 L of wastewater from the batik hand-drawn industry in Giriloyo. Cr(VI) was adsorbed using 0.5-3.0 g of activated carbon and with the contact time of 30-240 min. SEM revealed pores filled with adsorbate, whereas FTIR showed new bands at 600-500 cm−1 corresponding to the Cr-O spectral bond. The adsorption followed the Freundlich isotherm model, with a maximum capacity of 0.027 mg/g using 0.5 g at contact time of 30 min. The highest removal efficiency of (91%) was achieved with 3.0 g at 120 minutes contact time. These findings demonstrate that T. grandis activated carbon is an effective adsorbent for Cr(VI) removal, offering potential applications in sustainable batik wastewater treatment