Search Results (16,600)

Mobile cloud computing (MCC) is a technological paradigm for providing services to mobile device (MD) users. A compromised MD may cause harm to both its user and to other MCC customers. This study explores the use of machine learning (ML) models and stochastic methods for the protection of Android MDs connected to the mobile cloud. To test the validity and feasibility of the proposed models and methods, the study adopted a proof-of-concept approach and developed a prototype system named MINDPRESS. The static component of MINDPRES assesses the risk of the apps installed on the MD. It uses a device-based ML model for static feature analysis and a cloud-based stochastic risk evaluator. The device-based hybrid component of MINDPRES monitors app behavior in real time. It deploys two ML models and functions as an intrusion detection and prevention system (IDPS). The performance evaluation results of the prototype showed that the accuracy achieved by the methods for static and hybrid risk evaluation compared well with results reported in recent work. Power consumption data indicated that MINDPRES did not create an overload. This study contributes a feasible and scalable framework for building distributed systems for the protection of the data and devices of MCC customers. Full article

This article proposes a novel methodology aimed at streamlining the system’s development process. By examining existing state-of-the-art approaches and the capabilities inherent in Model-Based Systems Engineering (MBSE) tools, the article introduces a methodology centered around transforming a descriptive Systems Modeling Language (SysML) model into a digital twin. This virtual representation of the physical asset leverages real-time data and simulations to mirror its behavior and characteristics. When integrated with MBSE, this synergy allows for a comprehensive and dynamic approach, enhancing innovation by providing a holistic and adaptable framework for designing, analyzing, and optimizing complex systems throughout their lifecycle. The practical application of this Real-Time Communication and Data Acquisition (RT-CDA) methodology is implemented in a context and operational scenario of an electric unmanned autonomous vehicle employing both Software-in-the-Loop (SITL) and Hardware-in-the-Loop (HITL) approaches. The methodology empowers systems engineers to iteratively update and refine their system model’s fidelity based on real-world testing insights. The article specifically demonstrates the real-time communication capabilities achieved between an electric unmanned autonomous vehicle (a physical asset) and a descriptive (SysML) model, illustrating the real-time data aspect integral to the concept of a digital twin. This study serves as a foundation for future endeavors, envisioning real-time communication among virtual and physical models to construct comprehensive digital twins of complex systems to predict behavior and performance. Full article

This research aimed to evaluate block freeze concentration (BFC) under different centrifugation conditions using response surface methodology to separate an extract from the ice fraction at three centrifugal-BFC (CBFC) cycles, obtaining in the final cycle a phenolic-rich extract. A Box–Behnken design was applied to optimize centrifugation variables, with efficiency of separation (η) selected as the response variable. The extracts were characterized in terms of physicochemical analysis, total and individual bioactive components, and antioxidant capacity. Optimal conditions (3600 rpm, 16 °C, and 14 min) resulted in η of 82%. Thus, from infusion to final cycle, the solids, total polyphenol and flavonoid contents, and antioxidant capacity exhibited from 1.81 to 6.5% (w/w) and 2.5 to 8.7 (°Brix), 0.72 to 12.2 mg gallic acid equivalents/mL, 0.83 to 13.7 mg catequin equivalents /mL, 2.8 to 31.2 μmol trolox equivalents/mL and 4.8 to 122.2 μmol trolox equivalents/mL, identifying by high-performance liquid chromatography that kaempferol, p-hydroxybenzoic, and transferulic acid presented the highest concentrations. The CBFC process has the potential as a non-thermal concentration process to preserve many bioactive compounds, facilitating the production of concentrated fractions with high biological value, where the extracts obtained by BFC are a novel solution for medicinal, pharmaceutical, and food applications. Full article

Multiple input-multiple output (MIMO) is a key enabling technology for the next generation of wireless communication systems. However, one of the main challenges in the implementation of MIMO system is the complexity of the detectors when the number of antennas increases. This aspect will be crucial in the implementation of future massive MIMO systems. A flexible design can offer a convenient tradeoff between detection complexity and bit error rate (BER). Deep learning (DL) has emerged as an efficient method for solving optimization problems in different areas. In MIMO communication systems, neural networks can provide efficient and innovative solutions. This paper presents an efficient DL-based signal detection strategy for MIMO communication systems. More specifically, a preprocessing stage is added to label the input signals. The labeling scheme provides more information about the transmitted symbols for better training. Based on this strategy, two novel schemes are proposed and evaluated considering BER performance and detection complexity. The performance of the proposed schemes is compared with the conventional one-hot (OH) scheme and the optimal maximum likelihood (ML) criterion. The results show that the proposed OH per antenna (OHA) and direct symbol encoding (DSE) schemes reach a classification performance F1-score of 0.97. Both schemes present a lower complexity compared with the conventional OH and the ML schemes, used as references. On the other hand, the OHA and DSE schemes have losses of less than 1 dB and 2 dB in BER performance, respectively, compared to the OH scheme. The proposed strategy can be applied to adaptive systems where computational resources are limited. Full article

Common fig, or simply fig (Ficus carica L.), is one of the most ancient species originated and domesticated in the Mediterranean basin. The Italian fig germplasm consists of a large number of cultivars, more than 300. This number is approximate; there are many genotypes that are still poorly known and studied that may possess interesting agronomic traits, especially in terms of response to climate change. Therefore, it is extremely important to study and preserve agrobiodiversity, but more importantly to identify simple and rapid characterization methods to catalog “hidden” cultivated plants. In this study, geometric leaf morphometry was used to explore differences among fifteen Tuscan fig cultivars. In addition, the effectiveness of a machine learning (ML) algorithm to characterize cultivars was evaluated. The study analyzed two classes of cultivars, one of plants with predominantly three-lobed leaf shape, and one five-lobed. Thirty-three descriptors for the five-lobed and twenty-three for the three-lobed. Anova analysis showed statistically significant differences for all characters analyzed and allowed an initial characterization of the material. Then, Random Forest algorithm analysis was used to reduce the number of parameters to those most significant for classification. The results showed that machine learning-based techniques are a valid system for analyzing leaves of F. carica cultivars and interpreting significant differences in leaf parameters. Classification based on the Random Forest model allowed us to filter out the main descriptors that best differentiate cultivars from each other. Full article

As the key components of dietary carbohydrates, ensuring a balance between forage-neutral detergent fiber (FNDF) and rumen-degradable starch (RDS) is essential for ruminant health. Eight male Hu sheep equipped with rumen cannulas were randomly divided into four groups based on dietary FNDF content: low FNDF (L-FNDF, 6.08%), middle low FNDF (ML-FNDF, 9.47%), middle high FNDF (MH-FNDF, 12.48%), and high FNDF (H-FNDF, 15.68%), while the RDS levels (15.65% of DM on average) were similar among the four groups. A replicated 4 × 4 Latin square design was employed in this study. The results indicated that mean and minimum ruminal pH increased linearly with increasing dietary FNDF content, while the duration and area of pH under 5.8 and 5.6, along with the acidosis index, reduced linearly (p ≤ 0.002). There were no differences between the MH-FNDF group and the H-FNDF group in these indicators (p > 0.05). The molar proportions of acetate, butyrate, isobutyrate, and isovalerate, as well as the acetate-to-propionate ratio, increased linearly, while propionate and valerate molar proportions and lactate concentration displayed a linear decrease with increasing FNDF content in the diet (p < 0.001). Increasing dietary FNDF content extended ruminating and chewing time while enhancing ruminal microbial diversity, promoting the proliferation of Fibrobacterota and Butyrivibrio in the rumen, and improving fiber degradability (p < 0.05). When the dietary FNDF content exceeded 12.48%, no effects of FNDF on acetate to propionate ratio and fiber utilization were observed (p > 0.05). The results suggest that augmenting FNDF content in the PTMR can reshape ruminal fermentation towards acetate production and promote rumination to enhance ruminal pH, thereby alleviating the risk of ruminal acidosis. When the RDS content in the PTMR was 15.57%, an FNDF content of 12.48% was optimal for maintaining stable ruminal function in sheep, and the recommended ratio of FNDF to RDS was 0.8. Full article

Candida species, normally part of the healthy human flora, can cause severe opportunistic infections when their population increases. This risk is even greater in immunocompromised individuals. Women using intrauterine contraceptive devices (IUDs) are at higher risk for IUD-associated vulvovaginal candidiasis (VVC) because the device provides a surface for biofilm formation. This biofilm formation allows the normal flora to become opportunistic pathogens, leading to symptoms of VVC such as hemorrhage, pelvic pain, inflammation, itching and discharge. VVC is often linked to IUD use, requiring the prompt removal of these devices for effective treatment. This study evaluated the activity of the arginine-rich peptide “protamine” against Candida albicans, Candida tropicalis and Candida krusei isolated from IUD users who had signs of VVC. The antimicrobial activity was measured using the agar disk diffusion and microbroth dilution methods to determine the minimum inhibitory concentration (MIC). The MIC values of protamine against C. albicans, C. tropicalis and C. krusei are 32 μg mL⁻¹, 64 μg mL⁻¹ and 256 μg mL⁻¹, respectively. The determined MIC of protamine was used for a biofilm inhibition assay by crystal violet staining. Protamine inhibited the biofilm formation of the VVC isolates, and its mechanisms were studied through scanning electron microscopy (SEM) and a reactive oxygen species (ROS) assay. The disruption of cell membranes and the induction of oxidative stress appear to be key mechanisms underlying its anti-candidal effects. The results from an in vitro assay support the potential use of protamine as an antibiofilm agent to coat IUDs in the future for protective purposes. Full article

Recent advancements in artificial intelligence (AI), particularly in algorithms and computing power, have led to the widespread adoption of AI techniques in various scientific and engineering disciplines. Among these, materials science has seen a significant transformation due to the availability of vast datasets, through which AI techniques, such as machine learning (ML) and deep learning (DL), can solve complex problems. One area where AI is proving to be highly impactful is in the design of high-performance Li-ion batteries (LIBs). The ability to accelerate the discovery of new materials with optimized structures using AI can potentially revolutionize the development of LIBs, which are important for energy storage and electric vehicle technologies. However, while there is growing interest in using AI to design LIBs, the application of AI to discover new electrolytic systems for LIBs needs more investigation. The gap in existing research lies in the lack of a comprehensive framework that integrates AI-driven techniques with the specific requirements for electrolyte development in LIBs. This research aims to fill this gap by reviewing the application of AI for discovering and designing new electrolytic systems for LIBs. In this study, we outlined the fundamental processes involved in applying AI to this domain, including data processing, feature engineering, model training, testing, and validation. We also discussed the quantitative evaluation of structure–property relationships in electrolytic systems, which is guided by AI methods. This work presents a novel approach to use AI for the accelerated discovery of LIB electrolytes, which has the potential to significantly enhance the performance and efficiency of next-generation battery technologies. Full article

The influence of organosilicon compounds on butyl sealant blends’ mechanical and processing properties was investigated, particularly under increased humidity conditions. The addition of (3-mercaptopropyl)trimethoxysilane (MPTES), (3-aminopropyl)triethoxysilane (APTES), vinyltrimethoxysilane (VTMOS), and (3-glycidoxypropyl)trimethoxysilane (GLYMO) to elastomeric blends containing butyl rubber (IIR) and polyisobutylene (PIB) was studied. Key rheological parameters, including Mooney viscosity and melt volume rate (MVR), along with mechanical attributes such as peel resistance and cone penetration, were evaluated. Results indicated that functionalized silanes enhance sealant cohesion when their functional groups interact with the matrix and form cross-links under humid conditions. The presence of unreacted silanes acts as a plasticizer, increasing MVR and reducing viscosity. A notable MVR increase, up to 109 mL/10 min, was observed for the APTES-10 system. The most significant mechanical property enhancements were observed in blends containing MPTES and APTES, resulting in increased cohesion and peel resistance. The findings of this research are of considerable practical relevance, demonstrating that the modification of rubber sealants with monofunctional silanes improves their cohesion, delamination resistance, and processability, thereby making these materials suitable for the production of more durable sealants. Full article

Directed energy deposition (DED) has progressively emerged as a highly promising technology for the rapid, cost-effective, and high-performance fabrication of hard-to-process metal components with shorter production cycles. Recognized as one of the most widely utilized metal additive manufacturing (AM) techniques, DED has found extensive applications in critical industrial sectors such as aerospace and aviation. Despite its potential, challenges such as inconsistent part quality and low process repeatability continue to restrict its broader adoption. The core issue underlying these challenges is the complex, dynamic nature of the DED process, which involves the coupling of multiple physical fields. Within this context, the molten pool plays a pivotal role, serving as a key carrier that encapsulates abundant process characteristic information. The dynamic characteristics of the molten pool are intrinsically linked to the final part quality and the repeatability of the process. Consequently, integrating machine learning (ML) methodologies into the monitoring framework can offer robust data-driven support for enhancing both product quality and process consistency. This paper provides a comprehensive review of the research advancements and prospective trends in the dynamic monitoring and control of molten pool characteristics within DED processes underpinned by machine learning techniques. The review is structured around five key areas: an overview and fundamental principles of DED technology, methods for process information sensing during part monitoring, approaches for dynamically monitoring molten pool characteristics, the primary challenges currently faced in intelligent monitoring systems, and the potential future directions for further research and development. Through this detailed examination, the paper aims to shed light on the pivotal role of intelligent monitoring systems in advancing DED technology, ultimately contributing to more reliable and repeatable additive manufacturing processes. Full article

Heart rate variability (HRV) is a non-invasive health and fitness indicator, and machine learning (ML) has emerged as a powerful tool for analysing large HRV datasets. This study aims to identify athletic characteristics using the HRV test and ML algorithms. Two models were developed: Model 1 (M1) classified athletes and non-athletes using 856 observations from high-performance athletes and 494 from non-athletes. Model 2 (M2) identified an individual soccer player within a team based on 105 observations from the player and 514 from other team members. Three ML algorithms were applied —Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Support Vector Machine (SVM)— and SHAP values were used to interpret the results. In M1, the SVM algorithm achieved the highest performance (accuracy = 0.84, ROC AUC = 0.91), while in M2 Random Forest performed best (accuracy = 0.92, ROC AUC = 0.94). Based on these results, we propose an athleticism index and a soccer identification index derived from HRV data. The findings suggest that ML algorithms, such as SVM and RF, can effectively generate indices based on HRV for identifying individuals with athletic characteristics or distinguishing athletes with specific sports profiles. These insights underscore the importance of integrating HRV assessments systematically into training regimens for enhanced athletic evaluation. Full article

The rapid advancement of technology has increased the complexity of cyber fraud, presenting a growing challenge for the banking sector to efficiently detect fraudulent credit card transactions. Conventional detection approaches face challenges in adapting to the continuously evolving tactics of fraudsters. This study addresses these limitations by proposing an innovative hybrid model that integrates Machine Learning (ML) and Deep Learning (DL) techniques through a stacking ensemble and resampling strategies. The hybrid model leverages ML techniques including Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), eXtreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost), and Logistic Regression (LR) alongside DL techniques such as Convolutional Neural Network (CNN) and Bidirectional Long Short-Term Memory Network (BiLSTM) with attention mechanisms. By utilising the stacking ensemble method, the model consolidates predictions from multiple base models, resulting in improved predictive accuracy compared to individual models. The methodology incorporates robust data pre-processing techniques. Experimental evaluations demonstrate the superior performance of the hybrid ML+DL model, particularly in handling class imbalances and achieving a high F1 score, achieving an F1 score of 94.63%. This result underscores the effectiveness of the proposed model in delivering reliable cyber fraud detection, highlighting its potential to enhance financial transaction security. Full article

In lactating ewes, energy demand increases for milk production, reserve mobilizations, and body weight maintenance. For reconversion to energy, ruminants require ruminal propionate production because it is the most predominant substrate for gluconeogenesis and one of the most relevant pathways since it allows an adequate supply of glucose. Calcium propionate supplementation is an alternative to increase glucose production by an external additive. Thus, the objective was to evaluate the effect of calcium propionate (CaPr) on milk production and milk metabolomic profile on lactating ewes. Sixteen Rambouillet (65.3 ± 6.2 kg BW; three years old) were randomly assigned one of two experimental treatments: (a) basal diet without supplementation (CP/0S) and (b) basal diet + 30 g d⁻¹ of CaPr (CP/30S). The experimental period was from parturition day until day 60 (baby lamb weaning). A completely randomized design was used and analyzed with a mixed model. Initial and final lactating weight and milk production differed statistically (p < 0.05) between treatments. CP/30S led to differential changes (p < 0.05) in the lactation curve, showing significant milk production over eight-week measurements. Lactation peak (mL), maximum production (mL), and lactational persistency (d) were superior (p < 0.05) for supplemented ewes. An 11.4% variability was shown in a principal component analysis between treatments. For CP/0S, 63 bioactive compounds were detected, and 55 for CP/30S treatment. The metabolites detected in CP/0S showed that only fatty acid biosynthesis, biosynthesis of unsaturated fatty acids, and fatty acid elongation pathways were affected (p < 0.05) in milk. However, for CP/30S, metabolic pathways related (p < 0.05) were fatty acid biosynthesis, biosynthesis of unsaturated fatty acids, fatty acid elongation, phenylalanine metabolism, and steroid metabolism in milk samples. Calcium propionate supplementation increases milk performance and lactation persistency-induced changes in specific metabolic milk production pathways. Full article

This study introduces a framework that leverages the synergistic potential of Virtual Reality (VR) and Machine Learning (ML) to enhance graphical modeling in engineering and architectural design. Traditional clash detection methods in Building Information Modeling (BIM) systems are predominantly reactive, identifying discrepancies only after their occurrence, leading to costly and time-consuming design revisions. By integrating ML algorithms with VR-driven BIM, our approach proactively identifies and resolves clashes, as demonstrated across 28 diverse engineering projects. The results indicate a reduction in design clashes by 16% and iterative revisions by 15%, culminating in a 12% decrease in overall project timelines. This research underscores the transformative impact of combining VR and ML on additive manufacturing (AM) workflows, significantly improving efficiency and reducing the iterative nature of traditional methods. The findings highlight the framework’s scalability and adaptability, promising substantial advancements in engineering and architecture practices. Full article

Academic publications from the Web of Science Core Collection on “e-business” and “artificial intelligence” (AI) are investigated to reveal the role of AI, extract latent themes and identify potential research topics. The proposed methodology includes relevant graphical representations (trends, co-occurrence networks, Sankey diagrams), sentiment analyses and latent topics identification. A renewed interest in these publications is evident post-2018, with a sharp increase in publications around 2020 that can be attributed to the COVID-19 pandemic. Chinese institutions dominate the collaboration network in e-business and AI. Keywords such as “business transformation”, “business value” and “e-business strategy” are prominent, contributing significantly to areas like “Operations Research & Management Science”. Additionally, the keyword “e-agribusiness” recently appears connected to “Environmental Sciences & Ecology”, indicating the application of e-business principles in sustainable practices. Although three sentiment analysis methods broadly agree on key trends, such as the rise in positive sentiment over time and the dominance of neutral sentiment, they differ in detail and focus. Custom analysis reveals more pronounced fluctuations, whereas VADER and TextBlob present steadier and more subdued patterns. Four well-balanced topics are identified with a coherence score of 0.66 using Latent Dirichlet Allocation, which is a probabilistic generative model designed to uncover hidden topics in large text corpora: Topic 1 (29.8%) highlights data-driven decision-making in e-business, focusing on AI, information sharing and technology-enabled business processes. Topic 2 (28.1%) explores AI and Machine Learning (ML) in web-based business, emphasizing customer service, innovation and workflow optimization. Topic 3 (23.6%) focuses on analytical methods for decision-making, using data modeling to enhance strategies, processes and sustainability. Topic 4 (18.5%) examines the semantic web, leveraging ontologies and knowledge systems to improve intelligent systems and web platforms. New pathways such as voice assistance, augmented reality and dynamic marketplaces could further enhance e-business strategies. Full article