The increased traffic and heterogeneity in the 5G/B5G network require efficient radio resource management (RRM). However, the existing methods, such as GA and PSO, have poor convergence speed and require proper initial conditions. Additionally, learning-based methods have high computational complexity. Hence, in this paper, a novel AI-assisted Hybrid Channel Allocation (AI–HCA) framework is proposed by using a support vector regression (SVR)-based predictive initialization method and GA-PSO optimization. Simulation results using the 3GPP UMa channel model show that the proposed method has a 28% reduction in call blocking probability (CBP), a 15-25% enhancement in spectral efficiency (SE), and a 18-25% enhancement in energy efficiency (EE) with a faster convergence speed (35-40 iterations) than the existing methods. Additionally, the proposed method has been validated using the ANOVA test (p < 0.05) to confirm the significant improvements. Hence, the proposed AI-assisted Hybrid Channel Allocation framework has the potential to provide a low-complexity, robust, and scalable solution for intelligent radio resource management in 5G/B5G networks.
Palliative care is being influenced by artificial intelligence, especially when it comes to data-related aspects. In this context, care can be enhanced in terms of the quality of its results and the level of its efficiency, particularly with the help of technological tools such as artificial intelligence, which is capable of managing different types of information in the field of health care. Such characteristics have the potential to improve the quality of patient care while at the same time reducing the workload of healthcare professionals in palliative care. However, there are considerable ethical issues that need to be addressed with regard to the use of AI in this particular field to ensure that the benefits of the technological advancements are in the best interests of the patients and their human rights. The purpose of this review is to assess the current applications of artificial intelligence in palliative care and critically discuss the benefits in light of emerging ethical issues, particularly in low-resource settings. Some of the issues addressed in the literature include those of bias, data privacy, issues of consent, and the risk of depersonalization, all in the context of the principles of justice, beneficence, non-maleficence, and autonomy. Similarly, the review also identifies the increased challenges associated with the environments, which can be defined by the lack of infrastructure and/or regulation, thereby increasing the aforementioned issues. To address the aforementioned issues, the review suggests the need to develop policy, transparency through explainable AI, ethical auditing, and collaboration to ensure the safe usage of AI in palliative care.
The rising prevalence of obfuscated malware poses a critical threat to network security, undermining traditional detection methods and jeopardizing data integrity and system reliability in an increasingly connected world. This growing danger highlights the urgent need for advanced solutions to protect against evolving cyber risks. This research intro-duces a novel framework to enhance malware detection, employing a hybrid architecture that integrates spatial and temporal analysis with attention mechanisms. The approach leverages a large dataset subsample, focusing on key feature selection and augmentation to improve robustness against evasion techniques. This innovative framework offers a significant advancement in identifying malicious network traffic, addressing a vital gap in current security practices. The result demonstrates the model’s effectiveness, achieving an accuracy of 97.22%, underscoring its potential to strengthen defenses against sophisticated threats. Future efforts could further refine its performance, reinforcing its role in safeguarding network environments.
The rapid digitization of human activities has intensified reliance on internet-based platforms, creating fertile ground for cybercriminal exploits such as phishing. Despite advancements in detection mechanisms, phishing attacks continue to evolve, leveraging sophisticated visual mimicry to deceive users. This paper proposes a robust vision-based phishing detection system using ensemble deep learning to analyse webpage screenshots. The framework integrates transfer learning with pre-trained VGG16 and DenseNet121 models, extracting complementary low-level texture features (edges, gradients) and high-level hierarchical patterns (logos, layouts). These features are fused through a custom classifier with dropout regularization to mitigate overfitting. A balanced dataset of 3,000 webpage screenshots is curated augmented with horizontal flips, shear, and brightness adjustments to enhance generalizability. 5-fold cross validation techniques are used, where dataset is divided into 5 equal parts, where each one-fold served once as the test set while the remaining four used for training. The model achieves a training accuracy of 93.85% and testing accuracy of 89.00%, with a ROC-AUC score of 95.12% demonstrating strong separability between classes. Key contributions include a novel fusion of VGG16 and DenseNet121 for phishing detection, a lightweight architecture optimized for real-time inference (0.3 seconds per image), and a publicly available dataset to foster reproducibility. Experimental results surpass traditional HOG (Histogram of oriented gradients) based methods by 15% in F1-score and single-model baselines by 6% in accuracy. This work underscores the efficacy of ensemble deep learning in combating visually deceptive phishing attacks, offering a deployable solution for browser plugins or email security systems.
Electric vehicle (EV) uptake in Indonesia remains markedly below policy benchmarks. This study applies the Unified Theory of Acceptance and Use of Technology version 3 (UTAUT3), an extension of UTAUT2 that incorporates personal innovativeness as an additional construct to examine its impact on both behavioral intention and actual EV adoption within the Indonesian context. Unlike studies that typically survey the general public, this study focuses on actual EV users and owners, providing more representative and responsive insights into real-world EV usage. A total of 208 respondents participated, with 135 from the Jabodetabek area and 73 from Surabaya. The UTAUT3 framework explains 60.7% of the variance in behavioral intention and 62.5% in actual EV adoption in Indonesia. The results indicate that facilitating conditions, habit, personal innovativeness, and behavioral intention, significantly influence EV use behavior. Behavioral intention is positively influenced by hedonic motivation, price value, performance expectancy, and personal innovativeness, while social influence exerts a significant negative effect. Although facilitating conditions and habit do not significantly affect intention, they directly impact adoption behavior. These findings provide strategic implications for industry actors and policymakers aiming to accelerate EV adoption in the Indonesian market.
The review article demonstrates the importance of power electronics in propelling the increase of renewables and increasing the efficiency of contemporary electrical grids. In particular, it highlights the need for bidirectional converters for effective energy management in vehicle-to-grid (V2G) systems with two-way power transfer. The paper examines the series of complications offered by the significantly large integration of power electronic converter-interfaced renewables, such as those related to system steadiness, power quality, and resignation bottlenecks. Also, it recognizes a mix of new technologies, such as wide bandgap (WBG) semiconductors like silicon carbide (SiC) and gallium nitride (GaN), that deliver superior performance compared to traditional siliconbased devices. The review paper also covers the current trends in power electronics, control strategies, and energy storage systems using electric vehicles/hybrid systems. A comprehensive overview of the field has been presented by discussing the dynamic interaction between technological advancement and regulation frameworks, which outlines a broad view of sustainable energy systems in terms of power electronics. Furthermore, it highlights once more the importance of breaking down other barriers so that the much-needed transitions to low-carbon/highrenewable energy grids can take place and an affordable and reliable yet also sustainable and resilient future can be ensured.
Due to limited initiatives and support, the transition to green manufacturing among small and medium enterprises (SMEs) in developing countries remains challenging. Through a critical literature review and case study of the Indonesian vermicelli industry, this study highlights the importance of collaboration and strong policies to address this. Other findings also reveal that the energy intensity and emissions in the sector exceed best practice benchmarks, with the cooking and drying stages contributing significantly to emissions due to high energy consumption. Lack of government regulations, company awareness, and understanding of applicable practices and technologies creates challenges in green manufacturing. The study has utilized a multifaceted approach, including energy efficiency, green technologies, and renewable solutions like hybrid solar dryers. The study helps SMEs in developing countries implement green manufacturing and sustainable practices by providing empirical data and energy-emission benchmarks.
In this research, extraction of polysaccharide compounds from Cactus (Opuntia Ficus Indica) leaves in both fresh and dehydrated condition by solvent precipitation method using three types of water, acid (HCL) and NaOH. Before extraction, the physicochemical properties were examined to determine optimum yield (%) of extracted polysaccharide by optimization of Box-Behnken Design (BBD) of response surface methodology (RSM). The polysaccharide-based acrylamide hydrogel was prepared by free radical polymerization. The functional and structural characterization was done by FTIR, XRD and SEM for examination of extracted polysaccharide as raw polymer backbone in hydrogel preparation and prepared hydrogel as adsorbent for metal removal. From physicochemical analysis dehydrated Cactus had high viscosity of 4.0 cP and acidic nature around pH 5.0 with high carbohydrate, ash, fiber content with low moisture content. Optimum yield % was 3.82% for fresh and 19.75% for dehydrated condition. They have the same polar water group such as hydroxyl (OH), methyl (CH3), carboxyl (COOH), amino (amine) NH2 on the backbone of polymers in functional analysis. According to XRD and SEM, the result amorphous structure with polysaccharide nature of the bloom distribution surface was found in validated dehydrated CtG extraction. The amorphous nature with crosslinked pore structure surface of hydrogel was examined from XRD and FTIR results. From functional analysis, hydrophilic group; hydroxyl (OH), methyl (CH3), carboxyl (COOH), amino (amine) NH2 that provide the formation of three-dimensional network structure in prepared hydrogel with the ability to adsorb metal cation.
Bimetallic catalysts were prepared by deposition of 1% of loading of (Pt-Co) onto SBA 15 support using heterogeneous reaction for enhancing hydroisomerization and promoting hydrocracking for the decomposition of n-heptane. The catalytic performance of the catalysts was subsequently evaluated thereafter. Characterization of the prepared catalysts was done by various techniques, which include XRD, FTIR, BETBJH, SEM/EDS, and TEM. The temperature range studied for this catalytic activity was 250 to 400 °C, at atmospheric pressure in a plug-flow reactor. The experimental setup kept a very rigorous control over the temperature, flow rate for the reactants, and pressure, while gas chromatographic techniques served for analysis of the components exiting the reactor to determine product composition and concentration. Investigation results showed that the Pt-Co (1%)/SBA-15 catalyst was rather active and converted around 65% of all n-heptane presented in the reaction mixture. At the same time, the selectivity towards isomerization was not that high, which means much more detailed investigations and research are needed in this direction in the future to increase the efficiency and effectiveness of the catalytic process.
Low voltage Direct Current (LVDC) microgrids have been extensively used as a hopeful technology with high access to distributed energy resources (DERs) and DC loads, due to the absence of multiple conversions. A high amplitude Fault current together with the vulnerability of electronic devices, poses a great challenge to the line protection in LVDC microgrids. Electromechanical switches are slow and take long tripping time to clear the short circuit fault. Power electronic circuit breakers are used to achieve fast fault interruption recent progress in areas like machine learning and natural language processing have affected nearly every assiduity and area of scientific exploration, including electrical engineering. This paper proposes an Artificial Neural Network protection strategy with a Solid-State circuit breaker based bi-directional short circuit current blocker in series with bus for the protection of DC bus in LVDC Microgrid. Artificial neural network (ANN) technique is used for optimization of fault clearing time within a millisecond.
The proportional systems in vernacular architecture often exist without formal written codification, raising questions about their transmission across generations. Bubungan Tinggi House which is a replica of the Banjar Kingdom palace was built by Banjarese merchants and aristocrats from the 16th to the 19th centuries, showing a uniform aesthetic quality, which indicates a regularity in the proportions of its form. However, no historical written manuals detailing these proportions are known to exist. To prove the existence of regularity in the Bubungan Tinggi House proportion system, research was conducted by taking 7 Bubungan Tinggi House samples spread across South Kalimantan province, with varying sizes and built over different time periods. This study aims to investigate the hypothesis that the proportional system of the Bubungan Tinggi house functions as a set of "unwritten rules" transmitted orally and through practice. The methodology combines (1) a quantitative geometric analysis of seven Bubungan Tinggi houses to empirically confirm the existence and nature of the proportional system, and (2) a qualitative ethnographic approach through semi-structured interviews with community elders and descendants of traditional builders to explore the transmission mechanism. The ratio numbers obtained are single-digit integer numbers that are easy to apply, where the ratio numbers are: 1:1, 1:2, 1:4, 1:8, 2:5, 3:2, 3:4, 3: 5/5:3, 1:6, 6:5, and 9:20. The qualitative findings reveal that this knowledge was not articulated mathematically but transmitted through a master-apprentice model, guided by an intuitive aesthetic sensitivity (intuitive/sensory knowledge). This study concludes that the Bubungan Tinggi's proportional system is a form of tacit knowledge, embedded in cultural practice rather than textual documentation. These findings have significant implications for the conservation of architectural heritage, emphasizing the need to preserve not just the physical artifact but also the intangible craft traditions that produce it.
This study introduces a novel approach for early dyslexia detection in children through automated handwriting analysis, integrating a hybrid CNN-BiLSTM-CTC architecture with personalized learning strategies. Our method combines a custom CNN-BiLSTM-CTC model with tailored educational interventions to support dyslexic learners. We analyzed children's handwritten text images in English, collected from specially designed tests involving word rewriting, sentence reconstruction, and paragraph composition, particularly challenging tasks for dyslexic individuals. Notably, our CNN-BiLSTM-CTC model achieved the best result with an accuracy of 97.67%, outperforming other architectures. Compared to pre-trained models like EfficientNetB7, DenseNet121, and MobileNetV2, our custom CNN-BiLSTM-CTC model demonstrated superior performance. Key contributions include the development of a novel CNN-BiLSTM-CTC architecture for handwriting analysis and the integration of personalized learning strategies to enhance educational outcomes. By facilitating earlier and more accurate detection, this approach can significantly improve educational support for children at risk of dyslexia. Future research will focus on expanding the dataset and conducting longitudinal studies to assess the long-term impact. Customization and CNN-BiLSTM hybridization together enhance performance by capturing subtle handwriting variations and integrating spatial with sequential learning.
Instead of examining several components that link planning and strategy implementation, researchers have focused on the alignment of IT and business, as well as the Strategic Information Systems Planning (SISP) process. The goal of this study is to examine several approaches that can close the gap between IT project planning and execution. Therefore, the primary objective of the proposed study is to examine SISP implementation strategies that can be theorized in the context of Veolia, a global multinational corporation (MNC). The primary aspects influencing the SISP for this study are the frequency of planning, the function of IT/business managers, assessment or measurement criteria, critical success factors, and contextual considerations. To achieve methodological rigour, they first carried out a pilot study, and then conducted semi structured one-on-one interviews with 15 managers at Veolia; the IT, business and functional managers. The purposive sampling technique was used in selecting the participants as they were chosen to provide both the technical and the operational approach with regards to IT planning and implementation. This qualitative study will require considering the organizational level of analysis to obtain information on the planning and implementation of SISP in the corporate environment of Veolia. Brauns and Clarke (2006) six-phase framework of thematic analysis was used to analyze the data obtained in the interview to guarantee systematic coding, theme development and interpretation to increase the strength of analysis. The data or respondents' open-ended comments were grouped into more general themes. The study's conclusions showed that preparing IT initiatives takes longer than executing them. The study's findings also point to a favorable correlation between the planning horizon and the authority granted to IT managers or executives. Additionally, there is a direct correlation between the lack of authority granted to users during the implementation phase and the implementation failures they report.
One of the by-products of the fermentation process that produces ethanol, which is mostly utilized in industry and food, is methanol. For methanol adsorption in batch operations, mixed amines modified (MCM-41) was utilized. A batch adsorption technique loaded with MCM-41 sorbent was used in the study to separate methanol from ethanol. This study used methanol at varying initial concentrations (40-80 mg/L). As a result, the effect of temperature, duration of adsorption, amount of adsorbent and initial concentration of pollutant in the field of ethanol alcohol purification using MCM-41 was investigated through the adsorption mechanism. In addition, first and pseudo-second order kinetic adsorption models have been used to remove methanol pollutants from MCM-41 adsorbents with the aim of analyzing surface adsorption mechanisms. Additionally, a comprehensive analysis of the surface adsorption mechanism has been done. The models of Temkin, Freundlich, and Langmuir isotherms were applied. Several mass transfer models (Weber and Morris, Bingham's liquid film and Burt's diffusion) have been investigated on porous adsorbents to gain a deeper insight into the mechanism of pollutant mass transfer during surface adsorption. This study demonstrates that methanol adsorption from ethanol using MCM-41 achieves optimal efficiency at 60°C and an adsorbent dose of 0.5 g after 90 min. The results show that the process is spontaneous and follows a pseudo-second-order kinetic model, suggesting great potential for ethanol purification in industrial processes. The results showed that a mesoporous adsorbent known as MCM-41 is able to efficiently extract methanol from ethanol solution.
A decade after its establishment, APASTI has emerged as a key regional mechanism for promoting science, technology and innovation as drivers of sustainable development, resilience, and shared prosperity in Southeast Asia. This editorial highlights the contributions of APASTI to policy alignment, knowledge exchange, research visibility, and collaborative network-building across ASEAN, while also acknowledging ongoing constraints related to governance, capacity, funding disparities, and uneven regional integration. It introduces a special issue of 14 papers published in AJSTD that collectively evaluate the implementation and impact of APASTI across diverse sectors, including geohazards, energy, materials, health, infrastructure, transport, and information systems. By positioning these papers within the broader agenda of inclusive and open scientific publishing, the editorial argues that accessible regional scholarship is essential for informing the next generation of STI policy and cooperation in ASEAN. Although APASTI has established an important base for regional transformation, its long-term success will depend on sustained collaboration, adaptive governance, and continued investment in equitable scientific capacity across member states.
This paper addresses the issues in the data organization logic of existing intelligent steel bar processing systems, which are centered on construction projects. It investigates product-oriented data organization methods and the key technologies of logical correspondence between steel bar product identification codes and physical entities. The paper analyzes the product-oriented data organization logic and presents the corresponding data organization architecture. It also examines the feature extraction of steel bar product information classification and coding, as well as the dynamic completeness of the coding rule system. The paper provides a construction method for steel bar product identification codes and designs partial information classification and coding rules for steel bar products. This study proposes an innovative approach to data sharing and product reuse in intelligent steel bar processing systems, while also delivering critical support for reducing management costs in the steel bar product supply chain and enhancing accounting accuracy.
The development of the poultry industry must consider the principle of circular economy. Therefore, this study aims to develop a method of fermenting poultry manure as a new, unconventional feed, which can support the model of circular farming. Poultry manure obtained from laying hens was mixed with cassava solid waste at a ratio of 7:3 to maintain proper moisture content. The mixed manure fermented for 14 days into a 30 kg silo in triplicate using different additives, namely Lactiplantibacillus plantarum FNCC 0020 at 1 x 105 cfu/g (LP), Bacillus cereus LS2B as proteolytic bacteria at 1 x 105 cfu/g (BC), multi-purpose microbes at 1% (MP), and a treatment without additives (CON) as the contol. The multi-purpose microbes contained lactic acid bacteria, fibrolytic, proteolytic, and amylolytic. The results showed that MP presented higher organic matter and gross energy with lower crude fiber after fermentation than CON (p < 0.05) in terms of chemical compositions. During fermentation, all microbial treatments caused lower pH and Escherichia coli (not detected), with high lactate production (p < 0.05). In addition, all microbial treatments improved the physical quality by reducing the aroma of manure after ensiling. In the rumen, MP had higher in vitro dry matter digestibility and total volatile fatty acid than CON (p < 0.05). Based on these results, the use of multi-purpose microbes can improve the chemical composition, fermentation, and digestibility of fermented manure, which can be applied as an unconventional feed for ruminants.
The transition toward renewable energy-dominated power systems have accentuated the intricacies of frequency control, necessitating advanced regulatory mechanisms. This investigation articulates a tri-zonal frequency stabilization approach, employing a hybridized controller that synergizes Fuzzy Fractional-Order PI and Tilt-Integral-Derivative methodologies. The uniqueness of this approach is further amplified by the deployment of the Snake Optimization algorithm for precise parameter tuning. Set within a conventional grid topology integrated with assorted renewable energy sources, the study evaluates the controller’s adaptability and resilience through a series of comprehensive scenario-based analyses.
The Water Quality Index (WQI) is an essential metric for evaluating the usability of surface water resources, particularly in ecologically sensitive and high-demand areas like the Panch Prayag belt of Uttarakhand, India. This region, comprising five major pilgrimage towns—Devaprayag, Nandprayag, Vishnuprayag, Karnaprayag, and Rudraprayag—faces seasonal fluctuations in water quality due to both natural and anthropogenic pressures. In this study, water samples were collected from 2021 to 2023 across pre-monsoon, monsoon, and post-monsoon seasons, and the WQI was computed using the Canadian Water Quality Index (CWQI 1.0). Results revealed that WQI values ranged from 36 to 45 across locations and seasons, with lower values during dry seasons due to increased contaminant concentrations. Regression analysis using ANOVA identified magnesium, chloride, nitrate, and fluoride as key pollutants significantly influencing WQI (F-values > 10 in some locations, with p-values < 0.05), with location-wise R² values ranging from 75.6% (Vishnuprayag) to 93.1% (Rudraprayag). Artificial Neural Network (ANN) models were employed for predictive analysis, achieving high accuracy with R² values exceeding 0.91 and Root Mean Square Error (RMSE) below 0.6. The ANN model demonstrated a strong ability to forecast WQI trends, reinforcing its potential for real-time water quality monitoring. The study provides critical insights into pollution sources and seasonal dynamics, supporting sustainable water resource management in this culturally and environmentally vital region.
