International Journal of Communication Networks and Information Security (IJCNIS)
Not a member yet
1021 research outputs found
Sort by
Revolutionizing English Language Curriculum Implementation: Elevating English Language Lessons through Teacher Development
No full text
Received: 17 Apr 2024
Accepted: 26 Aug 2024
This research explores the impact of professional development on the integration of technology in teaching English language in five Trust Schools (TS)in northern states, Peninsular Malaysia. The study aims to evaluate the nature and quality of professional development programs, identify challenges faced by teachers, and measure the overall impact on teaching practices and student learning outcomes. A qualitative approach is utilized, interviews, classroom observations, reflective entry and focus group discussions with English teachers, school administrators, and students. The study seeks to answer two primary research questions: 1) what challenges do English language teachers face in integrating technology to implement English language curriculum? And 2) how does professional development influence English language teachers’ digital literacy in elevating English language lessons?Initial findings suggest that while professional development programs exist, their effectiveness varies significantly based on factors such as frequency, content relevance, and support mechanisms. Teachers often face challenges including limited resources, lack of ongoing support, and varying levels of digital literacy. Despite these challenges, professional development has been found to enhance teachers’ confidence and competence in using technology, leading to more innovative teaching practices and improved student engagement. The study concludes with recommendations for improving professional development programs, emphasizing the importance of continuous training, tailored support, and collaborative learning communities. These findings aim to inform policymakers and educational stakeholders on strategies to enhance technology integration in teaching English, ultimately contributing to better educational outcomes in Perlis
I10 Index and Academics Visibility
No full textThe i10 index is a Google Scholar innovation that was introduced in 2011. Google Scholar is a sub-browser from Google that enables the search of scientific literature, such as journal articles, theses, books, etc. It was created to provide metrics and a brief overview of the number of publications to measure the productivity and academic impact of a researcher in terms of the number of publications that have been collected a minimum of 10 times by other researchers. In other words, the i10 index shows the number of publications by a researcher that have been cited a minimum of 10 times. For example, if a researcher has an i10 index of 20, that means the researcher has 20 publications that have each been cited a minimum of 10 times. This helps provide an overview of the researcher's impact and involvement in the scientific community
Impact of Welfare and Industrial Safety Management Practices on Employee Satisfaction: A Study Within Construction Units
No full textThis study delves into the influence of welfare and industrial safety management practices on employee satisfaction within construction units situated in the northern region of India. A comprehensive questionnaire, developed through a synthesis of literature review, pilot testing, and expert evaluation, was administered to gather responses from 475 employees across various job categories within 18 construction units, constituting a population size of 3850 individuals. Subsequent to data collection, a rigorous analysis was conducted, employing both descriptive and inferential methodologies. The findings underscore a robust positive correlation between the adoption of welfare and industrial safety management practices and the augmentation of employee satisfaction. Furthermore, the study illuminates the fostering of a favourable organizational attitude among employees and an increased inclination to remain with their current employers. Notably, these practices have significantly contributed to heightened safety awareness, bolstered accident prevention measures, enhanced safety performance, and cultivated a stronger sense of workplace safety. These insights underscore the pivotal role of efficacious welfare and industrial safety measures in nurturing a supportive work environment elevating employee satisfaction levels at construction sites. However, it is imperative to acknowledge the limitations of this study, including its geographical and industry-specific scope, as well as its reliance on self-reported data, which may introduce biases. Accordingly, future research endeavours are recommended to expand the investigation across diverse regions and sectors, employing longitudinal studies to elucidate the enduring impacts of these practices on enhancing employee satisfaction and fostering the overall growth of the construction industry
A Novel Deep Learning Lacunarity Texture Analysis System Using Mid-Point ROI Extraction Algorithm for Palmprint Recognition System
No full textThe palmprint comprises multiple unique patterns that are distinct in detecting human identity. There are numerous algorithms proposed by past researches for recognizing Two-dimensional Palmprint Region of Interest (2DPROI) images. In this research, an innovative Deep Learning Lacunarity Texture Analysis System (D2LTA) is developed for recognizing the accredited persons at higher recognition rate. To impart the D2LTA model, Two-dimensional palmprint hands’ ROI images are segmented using Mid-point ROI generation algorithm, produced a peculiar feature vector using lacunarity approach in a state-of-the-art manner, and then Deep Learning ConvNet classifier is proposed for D2LTA system to justify the accredit person. The key principle of the Mid-point ROI generation approach is to determine the perfect straight line on the center of the palm. Based on the straight line in the palm, determine the pixel values of the ROI’s rectangular box. To catch the perfect straight line, line mid-point method is used. To do this research, 2D-palm hands are procured from three different datasets such as BMPD, CASIA and IIT palm datasets and 2DPROI images are secured from PolyU, Hong Kong Polytechnic University, Hong Kong. The proposed model has been assessed with diverse dimensions to prove the acquirement 99.25% of higher precious authentication rate
Advanced Join Query Optimization Using Firefly and Reinforcement Learning Techniques on Tpc-H Dataset
No full textJoin query optimization is a critical component of database management systems (DBMS), significantly influencing their performance and efficiency. This study delves into advanced optimization techniques by employing the Firefly Algorithm and its hybrid integrations with Deep Q-Network (DQN) and Double Deep Q-Network (DDQN) methodologies. Utilizing the TPC-H benchmark dataset, we rigorously evaluate the efficacy of these algorithms in optimizing complex join queries. The Firefly Algorithm, inspired by the luminescent communication of fireflies, serves as a powerful metaheuristic optimization technique, adept at navigating vast search spaces. To augment this method, we incorporate reinforcement learning via DQN and DDQN, enhancing the algorithm's capability to balance exploration and exploitation during the optimization process. Our empirical analysis reveals substantial performance gains with the hybrid DQN-Firefly and DDQN-Firefly approaches compared to the standalone Firefly Algorithm. These findings underscore the potential of these hybrid methods for practical implementation in database management systems, promising improved query optimization and overall system performance
Power-Based Single Threshold Load Balancing Algorithm For Cloud Computing
No full textCloud Computing has become a compelling concept in today's IT industry, enabling the development and deployment of custom applications across various domains using a pay-as-you-go model. This model supports hosting pervasive applications from consumer, scientific, and business sectors. Efficient resource utilization in cloud environments is crucial to prevent wastage due to under-utilization and to avoid prolonged response times caused by over-utilization. A significant challenge in cloud computing is the scheduling of user requests—specifically, how to allocate resources effectively to ensure tasks are completed quickly while maximizing resource efficiency. Load balancing plays a pivotal role in resource provisioning for high-availability cloud solutions. In this context, a Single Threshold Power-Based Load Balancing Algorithm has been proposed for cloud computing environments. The algorithm's fitness function is based solely on bandwidth to select the optimal virtual machine (VM) for task execution. The algorithm's performance is evaluated using Cloud Reports, Cloud Analyst Simulators, and MATLAB, focusing on key metrics such as power consumption, simulation time, CPU utilization, and cloudlet processing per data center. Additionally, the algorithm's performance is compared against the Single Threshold Algorithm, tested under two broker policies: time-shared and space-shared
Leveraging Natural Language Processing for Automated Detection and Profiling Of Emerging Cyber Threats
No full textThe evolving landscape of cyber threats demands swift detection and mitigation strategies, as highlighted by the Log4j vulnerability incident. This context emphasizes the critical need for cybersecurity defense strategies to adapt quickly to the narrowing window between the disclosure of a vulnerability and its exploitation. The urgency of early threat detection is paramount to thwarting cybercriminal activities effectively. However, security analysts face significant challenges due to the vast volume of data and diverse information sources. To tackle this issue, we propose a novel framework that utilizes Twitter messages as event indicators and employs the MITRE ATT&CK framework for detailed threat characterization. The proposed framework comprises three main stages: identifying cyber threats and their associated nomenclature, profiling the intentions or goals of identified threats through a two-layered machine learning approach for filtering and classifying tweets, and generating alerts based on a comprehensive threat risk assessment. A key contribution of our framework is its ability to provide in-depth insights into the intentions or goals of identified threats, thereby enriching threat context and enhancing mitigation strategies. In experimental evaluations, our profiling stage demonstrated a high F1 score of 77% in accurately characterizing detected threats. This framework represents a notable advancement in threat detection and mitigation, offering a sophisticated yet efficient approach to addressing emerging cyber threats in an increasingly unpredictable digital environment
PERFORMANCE ANALYSIS OF NEUTRAL POINT CLAMPED MULTI-LEVEL INVERTER WITH DIFFERENT PWM TECHNIQUES
No full textThis paper presents a comprehensive performance analysis of a Neutral Point Clamped (NPC) Multi-Level Inverter using various Pulse Width Modulation (PWM) techniques. The NPC inverter is a widely adopted topology in medium to high-power applications due to its ability to deliver high-quality output with reduced Total Harmonic Distortion (THD). This study investigates the impact of different PWM techniques, including Sinusoidal PWM (SPWM), Space Vector PWM (SVPWM), and Level-Shifted PWM (LSPWM), on the performance of the NPC inverter. Through detailed simulations and experimental validations, we evaluate key performance metrics such as THD, efficiency, and output voltage quality under different load conditions. The results reveal that while all PWM techniques effectively reduce THD, SVPWM consistently achieves the lowest THD and highest efficiency across various operating conditions, making it the optimal choice for high-performance applications. SPWM, although easier to implement, demonstrates slightly higher THD, while LSPWM provides a balanced approach with moderate THD and efficiency. This paper's findings contribute to the optimization of NPC inverters for critical applications in renewable energy systems and industrial drives, providing valuable insights into the selection of PWM strategies to meet specific application requirements
Deep Transfer Learning for Masked Face Reconstruction and Hybrid DCNN-ELM Framework for Recognition
No full textFacial reconstruction has always been a pivotal aspect of medical and forensic science. The increasing use of face masks in recent years has posed new challenges, making traditional facial recognition techniques less effective. To address this, our research explored innovative methods for reconstructing faces from images obscured by masks. We focused on post mask face reconstruction and facial recognition using cutting-edge techniques. We assess the effectiveness of three key unmasking algorithms: edgeconnect (EC), gated convolution (GC), and hierarchical variational vector quantized autoencoders (HVQVAE). Using two synthetic face datasets, MaskedFace-CelebA and MaskedFace-CelebAHQ, we rigorously evaluate the quality of the reconstructed faces based on metrics such as the PSNR, SSIM, UIQI, and NCORR. Among these, the Gated Convolution algorithm stands out as the superior choice in terms of image quality. For facial recognition, we employ a novel hybrid framework that combines a deep convolutional neural network and an extreme learning machine (DCNN-ELM). We tested five classifiers (Vgg16, Vgg19, ResNet50, ResNet101, and ResNet152) in combination with ELM and a support vector machine (SVM). Our comprehensive ablation study revealed that ResNet152 combined with ELM achieved the best performance, with a facial recognition accuracy of 60.9%, suggesting that the reconstructed faces were of high quality. Our paper presents a novel approach to image classification utilizing five classifiers within the DCNN+ELM hybrid framework and provides a complete ablation study of these classifiers. This research underscores the importance of face reconstruction in the current field and its potential to enhance facial recognition techniques
Artificial Neural Network based Machine Learning Technique for Liver Disease Prediction
No full textLiver diseases, such as cirrhosis, hepatitis, and fatty liver disease, have become an increasing global health concern in recent decades. The rise in these conditions poses a serious threat to public health and places a significant strain on healthcare systems. Early detection is critical in managing liver diseases effectively, but the complexity of the contributing factors and the variability in patient data make accurate predictions challenging. Researchers are confronted with the task of analyzing vast and heterogeneous datasets containing key clinical and biochemical parameters, including liver function tests, demographics, and lifestyle factors. To address this challenge, sophisticated machine learning techniques are needed to uncover meaningful patterns in the data that can reliably predict the onset of liver diseases and assist in timely diagnosis and prevention. This study proposes an Artificial Neural Network (ANN)-based solution that leverages a comprehensive dataset encompassing critical parameters such as age, gender, total bilirubin, direct bilirubin, and alkaline phosphatase levels. By processing these inputs, the ANN model is designed to identify subtle patterns and make accurate predictions for early liver disease detection. Preliminary results indicate that this deep learning-based approach has the potential to outperform traditional diagnostic methods, offering a faster, more accurate tool for healthcare providers. The integration of machine learning into liver disease diagnosis not only enhances predictive accuracy but also holds the promise of revolutionizing how these diseases are detected and managed, ultimately improving patient outcomes and reducing healthcare burdens