Journal of Advanced Research Design

Hybrid Feature Selection Method using Novel Pasi-Luukka and Genetic Algorithm Method for Microarray Cancer Classification

2025-03-07T07:04:52+07:00

Deoxyribonucleic acid (DNA) microarray technology enables the simultaneous measurement of the expression level of numerous genes, thus enabling the identification of patterns in gene expression that may cause a disease or a particular biological process. The DNA microarray technology can identify cancer cells by analysing the gene expression difference between normal cells and cancer cells. However, due to the vast number of features in the DNA microarray, the feature selection method is required to identify the most relevant subset of microarray features for subsequent analysis. In this research paper, a novel hybrid feature selection method called Pasi Luukka-Genetic Algorithm + Support Vector Machine is introduced. This approach combines the strengths of filter and wrapper methods to effectively select features from eight (8) cancer datasets. The Pasi Luukka algorithm filters irrelevant features and reduces dimensionality. A metaheuristic-based feature selection, which is the Genetic Algorithm (GA), selects the optimum features from the filtered features. This paper evaluates the performance of the proposed method, and a comparison work is conducted against other existing hybrid feature selection methods in the literature. The evaluation considers accuracy metrics and the number of selected features using the same microarray datasets.

Development of Cockleshell Filter as Adsorbent in Palm Oil Mill Effluents

2025-03-25T13:05:35+07:00

The Malaysian palm oil industry has experienced significant growth, contributing to a substantial increase in processed waste, particularly palm oil mill effluent (POME). POME contains highly concentrated compositions that contribute to its brownish colour, posing environmental and health risks. The adsorption method, particularly using cockleshells (Anadara granosa), has emerged as a promising and environmentally friendly approach. The study aims to assess the performance of cockleshells in treating POME, focusing on reducing pollutants and colour to ensure compliance with effluent discharge standards for crude palm oil mills.

Morphological Descriptors as Tool for Characterization of Nuclear Pleomorphism in Breast Cancer

2025-03-25T08:27:23+07:00

In recent years, the advancement in molecular pathology and genetic analysis of cancerous tissues has significantly remarked an increase in objective and measurable data. However, the traditional approach of morphological analysis in pathology diagnosis remains subjective in comparison, despite the introduction of digital pathology aiding computer-aided diagnosis. Certain pathological grading features, such as nuclear pleomorphism in breast cancer, still depend on a pathologist's expertise and are largely non-quantitative. This study aimed to investigate morphological descriptors as key elements to characterize the qualitative description of nuclear pleomorphism in breast cancer, in line with the Nottingham Histopathology Grading (NHG) system. Four morphological descriptors were extracted from segmented nuclear cells, including area, minimum ferret diameter, minor axis length and perimeter and used to assign scores of 1 to 3 to characterize pleomorphic nuclei. The proposed method was validated using the support vector machine (SVM), achieving promising results with 95.0% and 92.0% in accuracy (Acc) and F1 score (F1), respectively. This study serves as a pilot investigation for the quantitative measurement of nuclear pleomorphism in breast cancer.

Optimizing AMI Control Centres through Machine Learning: A Review

2025-02-03T06:47:53+07:00

Modernizing electrical grids requires advanced metering infrastructure (AMI), which gives users and suppliers energy usage data and boosts grid efficiency. Utility smart meter operation centres monitor and analyse these benefits to maintain them. However, the massive data quantities make manual data management impractical. This paper discusses how data analytics and machine learning (ML) can automate and optimize this process to improve control centre decision-making. Our research examines global smart meter implementation and how ML helps operators with identifying problems, preventive maintenance, network selection and cybersecurity. These applications decrease manual labour, enhance accuracy and boost productivity. We also discuss recent AMI trends to help utilities, governments and regulators plan energy. This article shows ML's disruptive potential in smart meter management by focusing on network dependability, operational safety, maintenance optimization and cybersecurity. Our findings show how ML permits utilities to provide a seamless, robust and customer-centred experience, bolstering AMI as a modern electric grid basis.

Experimental Study of High Availability Cloud Learning Management System and Monitoring System Based on Grafana, Prometheus, and Telegram

2025-06-04T08:43:42+07:00

This paper discusses the optimization of server management for digital-based education services, especially Learning Management System (LMS) services in the learning process. The research was conducted at the data center of SMK Negeri 2 Yogyakarta which has successfully adopted the Learning Management System (LMS), but has not been optimal in its application. The results of the experimental study resulted in performance optimization efforts through the application of High Availability using the load balancing method with the Round Robin algorithm as a load balancing method using HAProxy as a load balancer on the web server, as well as the cluster method through Galera Cluster on the database to minimize failures on the LMS. In addition, this research successfully developed a monitoring system using Grafana and Prometheus to provide comprehensive visibility into the performance of the LMS cloud server. The system is equipped with real-time notifications via Telegram that allow server managers to respond quickly to any disruptions, thereby reducing downtime and increasing system reliability. Evaluation of the results of the implementation of cloud-based LMS with high availability and monitoring system based on Grafana, Prometheus, and Telegram at SMK Negeri 2 Yogyakarta proved to be effective in supporting the digital learning process which is expected to be a reference for other educational institutions that will adopt similar technology to improve the quality of learning services.

Review on the Stressors and the Psychosocial Factors Affecting Workers in Heavy Industries

2025-03-11T15:00:37+07:00

The issue addressed in this review is the significant impact of psychosocial factors on mental health within the global industrial sector, with a focus on mining, construction and the oil and gas industry. It delves into the pervasive influence of stressors, ranging from organizational functioning to physical conditions and how they affect workers in these industries. Specifically, workers in mining encounter challenges that affect both efficiency and safety, while the construction sector grapples with issues such as poor organizational structure. In the oil and gas industry, hazardous conditions and social isolation contribute significantly to worker stress. The study emphasizes the necessity for tailored interventions that take into account factors such as age, education and health conditions to address industry-specific stressors effectively. Furthermore, it highlights the global nature of workplace stressors and underscores the importance of holistic approaches for enhancing employee well-being. This review emphasizes the urgency of addressing these industry-specific stressors to promote mental health and improve overall well-being in the evolving landscape of global industrial work. Additionally, it aims to compare the listed factors and stressors among the three heavy industries: mining, construction and oil and gas.

Compressive Strength and Density in Fresh and Dried Cube Analysis of Autoclaved Aerated Concrete in Cooperating Glass Industrial Waste

2025-04-06T06:48:39+07:00

In the pursuit of sustainable construction materials, cooperating industrial waste into innovative building solutions has emerged as a promising way. Autoclaved Aerated Concrete (AAC) stands out as one such eco-friendly material renowned for its lightweight and insulating properties. The research highlighted that glass Industrial waste (GIW) had the potential to improve the strength and fire resistance of AAC due to its excellent properties of GIW in terms of physical, mechanical and thermal performances. The possibility of GIW was explored by including it in AAC concrete and analysing the effect of varying additions on the properties of AAC. This fundamental research proposed different percentages of GIW composition (5%, 10%, 15%, 20%, 25% and 35%) as a partial replacement for 70% (Sand+ Gypsum) mixed with different amounts (0.1% and 0.065%) of Aluminium paste. Then, the same percentages of the GIW composition (5%, 10%, 15%, 20%, 25% and 35%) were used as a partial replacement to reduce the silica content of 68% (Sand+ Gypsum) mixed with 0.1% Aluminium paste. All specimens underwent a steam curing process for 12 hours at a temperature of 180°C and a steam pressure of 13 bars in an autoclave machine to produce Autoclaved Aerated Concrete based on Glass Industrial Waste (AAC-GIW). These aspects were crucial in determining the optimum composition of glass waste that could influence the physical, mechanical and thermal properties of AAC-GIW. The optimum composition of GIW, sand, gypsum, lime, cement and aluminium paste as an expanding agent in AAC-GIW were the factors contributing to high strength. The compressive results were analysed in 3 different ratios (Ratio A, Ratio B and Ratio C) for fresh cubes and dried cubes, the work density and dry density. It is revealed that 15% GIW for ratio C is an optimum compressive strength at 2.43 MPa work density of 677 kg/m3 for fresh cubes as a sand replacement, in dried cube condition, 15% produces the highest compressive strength at 6.64 MPa of 661 kg/m3 for Ratio B. At different ratios, the most effective GIW content differed, emphasizing the requirement for modifications in AAC-GIW formulations specific to the AAC.

Revolutionizing Education: Exploring Technological Innovations, Opportunities and Challenges Across I0T, AR, VR, AI, LMS, ML, Gamification and Emerging

2025-01-27T11:35:54+07:00

Technological advancements are reshaping science and engineering education by fostering interactive, hands-on, and personalized learning experiences. This review examines the role of key technologies, including the Internet of Things (IoT), Augmented Reality (AR), Virtual Reality (VR), Artificial Intelligence (AI), Learning Management Systems (LMS), Gamification, 3D Printing, Blockchain, Cloud Computing, Robotics, Big Data, and machine learning (ML), in making a modern education. Bibliometric analysis was employed to identify patterns in publication growth, disciplinary focus, and geographical distribution, providing quantitative insights into the evolution of research on educational technologies. Each technology offers unique contributions, such as IoT-enabled smart classrooms, AR and VR for immersive visualizations, AI-driven analytics for personalized learning, and 3D printing for prototyping. The findings highlight the benefits of these innovations, including enhanced engagement, deeper understanding, and global access to resources. Challenges such as high costs, digital divides, inadequate infrastructure, and the need for educator training persist. Using a qualitative review approach, the study synthesizes recent research to provide insights for educators, policymakers, and researchers. It underscores the need for collaborative efforts to overcome barriers and maximize the potential of these technologies in science and engineering education. The results point to future directions, emphasizing the importance of equitable access, teacher training, and scalable solutions to ensure that technology continues to enhance educational outcomes.

Optimization of Handover Delay during Authentication

2025-03-21T09:41:30+07:00

In this work, the significant problem of optimal handover delay in NGN authentication was considered. This study reports an innovative Authentication, Authorisation and Accounting (AAA) architecture together with a far-reaching authentication efficiency enhancement across various networks based on the Universal Seamless Network Access Protocol (USNAP). This research employs a cloud-based AAA server and introduces the concept of a validity certificate (VC) to achieve fast user authentication. The performance of the proposed system is analysed via extensive simulations performed in NS-2. The results show a significant reduction in handover latency, with an average service time of 1.13 ms and a total handover delay of 77.47 ms, demonstrating that it is able to reduce the latency of previous approaches by roughly 13%. With this approach, the USNAP protocol can encode the type of network faster than current protocols and perform account verification. It is capable of scaling massively in our system with high service probability as the user needs grow. This work is helpful for securing light-weight and high-quality network signal authentication methods in NGNs and can be used seamlessly across various types of networking technologies.

Examining the Impact of Negative Bias Temperature Instability on the Performance of Domino Logic Circuits

2025-04-06T06:30:13+07:00

Negative Bias Temperature Instability (NBTI) poses a notable reliability concern in Metal-Oxide-Semiconductor Field-Effect Transistors (MOSFETs), causing an aging effect that alters threshold voltage and reduces drain current. This effect holds particular significance in sub-micrometre CMOS circuitry. This study focuses on assessing NBTI's impact on domino logic circuits, exploring various NBTI defect mechanisms like interface trap (Nit) and oxide trap (Not). Evaluations extend to NOR and NAND domino logic circuits, analysing delay and average power to gauge NBTI effects. The study employs the Predictive Technology Model (PTM) based on 32nm technology, coupled with the MOSRA model, to illustrate circuit dependability. Simulations involve varied stress temperatures, revealing a proportional degradation in delay with increasing temperature. Specifically, when Nit serves as the sole defect mechanism, the time exponent stands at 0.25, whereas Nit and Not together reduce this exponent to 0.167. Higher stress temperatures correlate with increased delay, reduced average power, and a shift in threshold voltage towards higher values over prolonged stress durations.

Design of Prototype for Metal Detecting Arduino Mobile Robot using Wireless Sensor Network Powered by Solar Energy

2025-03-11T13:59:50+07:00

In today’s modern technology world, there are several types of mobile robots which use different types of renewable energy systems. The advancement of modern technologies is increasing energy demand volume and is aimed at adopting alternative sources of energy. Renewable energy systems, such as solar power are the better choice to use in moving mobile robots. Robots are the fast-growing technology and communication mode to human in daily lives. Therefore, researchers have integrated solar power as an alternative energy source on mobile robots, which will solve the depletion of fossil fuels. Thus, a four wheels’ metal detection solar powered mobile robot was designed in these studies. The mobile robot is implemented using solar panels, Arduino Uno, metal detector sensor and solar power. The Android Uno was implemented in the development of this mobile robot as its primary brain and to communicate with all its components. In these studies, solar power metal detecting mobile robot is developed to detect the metal and the mobile robot is powered by solar energy and the robot control via wireless sensors network controller. Studies were also done on solar panel testing with various atmospheres and temperatures to accustom the product. The automated robot and the proposed design are compared in terms of the mobile robot's accessibility, metal detecting accuracy and stability of power source. The findings demonstrate that the mobile robot's proposed design can deliver faster metal detection with a more precise, easier in turning and highly dependable power supply.

Spine Tumor Segmentation using Deep Learning: A Review

2025-05-17T21:55:43+07:00

he field of medicine has been significantly impacted by technological advancements, particularly in digital imaging and image processing. These advancements have revolutionized early disease detection, computer-aided diagnosis, minimally invasive procedures, and image-guided surgeries. However, medical images, including spine tumors, often face challenges such as low contrast, noise, and artefacts, which impede accurate diagnosis. This paper reviews spine tumor image segmentation techniques utilizing Deep Learning (DL). It explores the crucial role of image segmentation in isolating specific anatomical structures, such as spine tumors, for precise diagnosis. DL has shown great potential in medical image segmentation, learning hierarchical features directly from raw data without manual feature engineering. The review highlights the significance of early spine tumor detection, classifies tumor types, and examines features of benign and malignant tumors. It emphasizes the role of accurate segmentation in improving surgical outcomes and advancing computer-aided diagnostic systems. Additionally, challenges in standard MRI protocols for distinguishing intradural from extradural tumor compartments are addressed, proposing advanced imaging techniques and DL models as solutions.