NPS Australia Submission System
  • 0

    No products in the cart.

Archives: Paper Submission

Intelligent Arduino based and analogue cocoa weighing combined scale system

Falsification of analogue weighing scales has gradually decreased cocoa production in Ghana. These falsifications are due to adjustments made to the currently used analogue scale system.
The study therefore proposes an intelligent Arduino based and analogue cocoa weighing combined scale system to replace the existing analogue weighing scale approach.
This design is highly recommended for cocoa buying companies to keep their staff in check and help increase cocoa production in Ghana.

Deployment of AI Models and a TDA Mapper Algorithm to Enhance Clinical Decision-Making

This research presents a novel Clinical Decision Support System (CDSS) that integrates advanced AI technologies, including a hybrid model for COVID-19 diagnosis and a custom GPT Assistant, into a scalable, real-world tool for resource-limited healthcare settings. The CDSS enhances the quality of care by providing accurate, timely diagnostics and decision support, while also improving response efficiency through real-time monitoring and predictive analytics. The system’s adaptability, supported by open-source platforms like R Shiny, and its potential for wide adoption, particularly after a planned pilot and impact evaluation, highlight its significance in advancing healthcare delivery where resources are scarce. The deployment of this AI-driven CDSS has the potential to transform clinical decision-making, improve healthcare delivery, and enhance the overall response to public health emergencies. The structured pilot and impact evaluation will provide critical insights into the system’s effectiveness in real-world settings, paving the way for broader adoption and sustained improvements in healthcare quality and efficiency.

Topological Machine Learning: Integrating Topological Data Analysis with Machine Learning to Enhance Breast Cancer Classification

Novel TML Approach: Introduces Topological Machine Learning (TML), integrating Topological Data Analysis (TDA) with Machine Learning (ML) to enhance breast cancer classification.

Improved Accuracy: Demonstrates that TML significantly improves classification accuracy on the Wisconsin Breast Cancer (WBCD) dataset compared to traditional methods and standalone ML techniques.

Feature Evaluation: Provides a detailed assessment of topological features such as cluster means, node features, and link features, highlighting their impact on model performance.

Practical Insights: Offers valuable insights into the integration of topological features for better diagnostic accuracy, with implications for improving breast cancer classification and patient outcomes.

Determining the Colliding Vehicle in Traffic Accidents Using Hybrid Machine Learning Models

In a world rife with vehicular accidents and traffic incidents, it is known that drivers are more likely than not to shift the blame in an accident rather than admit it. Other than that, there is a noticeable lack of models in the academic sector that allow neural networks to differentiate colliding vehicles from one another and are instead fixated on tracking and detecting traffic accidents as a whole. As such, the researchers propose a way of detecting colliding vehicles and classifying both vehicles as either the ‘colliding’ vehicle or the ‘collided’ vehicle. The processes in this machine learning pipeline are split into three main parts: crash detection—to which the model would use a crash detection algorithm; footage tracking—of which the model would utilise DeepSORT; and lastly a colliding vehicle classification algorithm that uses Gated Recurrent Units (GRUs), all of which will be combined to form a novel machine learning pipeline. The model exhibits very mixed performances when detecting both Vehicle 1 and Vehicle 2 in our testing phase. When detecting Vehicle 1, the model provides a very poor recall and F1-score, meanwhile the detection of Vehicle 2 exhibits a decent amount of precision, recall, and F1-score. Overall, the model provides an accuracy of 42% with a macro average precision of 0.45, a macro average recall of 0.29, and a macro F1-score of about 0.30.

Community Battery System Sizing To Maximize Financial Returns to the Prosumers in PV-Rich Neighborhood

The paper extends the knowledge on Community battery systems and sustainable energy.

Utilizing Caesium-based Vacancy-formed Materials For All-Perovskite Tandem Solar Cells: Photovoltaic Evaluation Using SCAPS 1-D

In this study, we reported the simulation study of lead-free all-perovskite tandem solar cell comprised of varied vacancy formation perovskite structure as wide bandgap absorber layer for FAMASnGeI3. The observations revealed that all-perovskite tandem solar cell of FTO/ZnO; 400 nm/Cs2AgBiBr6; 600 nm/FAMASnGeI3; 200 nm/Cu2O; 100 nm/Au achieved a notable PCE of 22.63 % at an operating temperature of 300 K with Jsc; 27.09 mA/cm2 Voc; 1.10 V and FF; 75.68 %. These findings suggests that this cell has strong potential in converting sunlight into electrical energy and indirectly will contribute to the advancement of environmentally friendly and high-performance solar cells, promoting the broader adoption of renewable energy technologies.

Impacts of Uncoordinated Electric Ferry Charging on Distribution Network

This study examines the potential effects of uncoordinated EF (electric ferry) charging on local distribution networks, focusing on Gladstone Marina in Queensland, Australia. Using OpenDSS software, power flow analysis assesses the simulated network with BESSs (Battery Energy Storage Systems) which represent proposed charging stations.

A Channel Selection Strategy for Energy Harvesting in Cognitive Radio IoT Networks

formulated an optimal channel selection strategy based on the combination of reliable reputation model and multiarmed bandit (MAB) problem to determine an optimal channel selection policy for the SU’s. With the main goal to maximize the SUs harvested RF energy from the PUs channels during transmission.

Automated Leak Detection in Drip Irrigation Systems using RGB and Thermal Sensor Fusion

Water leaks are a common issue in surface drip irrigation systems. Visual inspection of irrigation pipelines by humans is the most prevalent method for leak detection. However, this approach is costly and labour-intensive due to the need for frequent on-site visits. This paper describes an AI based sensor fusion algorithm to automatically detect leaks along the drip lines using RGB and thermal images collected from a low-cost ground vision system. The proposed algorithm was tested using images collected from vineyard under various light conditions. Results indicated that proposed sensor fusion detection algorithm is accurate and efficient.

Is Benford’s Law Based Detectors Effective for GAI Generated Images?

Benford’s Law is used in image forensics. We tested GAI generated images to see whether BL is able to detect artificially generated images. The experiments show that only about 60\% of the images can be detected using a simple similarity threshold.

ActJOLO: Action Recognition Guided by Actionlets Using Joint Lightweight Optical Flow Information

In this study, we propose a novel method, named ActJOLO, which builds upon the existing JOLO model by incorporating an advanced self-supervised learning technique as an upstream guide for posture recognition. Our approach emphasizes the analysis of high-intensity motion features within the human body, thereby enhancing the efficiency of action modeling.
Experimental results on the NTU RGB+D dataset demonstrate that our framework improves processing speed compared to the original model, while maintaining high ccuracy. This work offers a new perspective on skeleton-based human action recognition and highlights its potential for deployment on low-performance processors.

YuihaFS: Creating Versions for Each File in the File System

We propose a file system with novel snapshot function, called YuihaFS. The proposed file system reduces the disk usage for the differential data by allowing users and applications to select a file for creating snapshots. With this new property, YuihaFS can reduce the amount of differential data for snapshots by avoiding creating unnecessary snapshots.

Incorporating Human Intuitions into Data Augmentation to Detect Concentration on Conversation

This study proposes a data expansion method to classify the excitement of conversation. This study incorporates human intuitions for conversation excitement into data augmentation. Quantification of the human intuitions would efficiently assign correct labels to the data set generated by data augmentation. The pandemic of the new coronavirus has resulted in a loss of communication opportunities. We have lost opportunities that are important to form good relationships. A deep learning model to discriminate conversation excitement would contribute to increasing such important opportunities. However, training and using models to solve real-world problems requires a lot of data. There are many cases where sufficient data cannot be collected to train a model. In such cases, data augmentation is the most promising solution. We should pay attention to the point that effective data augmentation methods vary depending on the type and characteristics of the data. This study experimentally collects conversational data. It performs data augmentation on the conversational data. It creates datasets by similarity and trains multiple models. Comparing the accuracy of these models verifies the effectiveness of incorporating human intuitions into data augmentation. The paper discusses what kind of data augmentation technique works well to generate realistic conversation data with augmentation.

Enhancing Interpretability of Skin Lesion Classification using Grad-CAM and Weighted Grad-CAM

Introduced a new novel weighted Grad-CAM to give more insights into how the CNN Models give their result, based on the HAM10000 dataset for skin cancer classification and Interpretability.

Machine Learning in Phase of Flight Detection

The study integrates a thorough investigation of machine learning approaches with unsupervised clustering algorithms to advance the area of flight phase identification in aviation. We initiate our research with a comprehensive assessment of the literature on machine learning applications for flight phase detection, both scientometrically and conceptually. This review looks at how machine learning techniques have developed to meet theoretical and practical goals in this field, as well as publishing trends, relevant papers and top research institutes.
A significant contribution of our work is the application of several unsupervised clustering algorithms to flight data, which is simulated using Monte Carlo Simulation. This allows us to identify crucial flying phases, namely the straight and level phase and the turn phase. Through the performance evaluation of these techniques, we find that among the models examined, the Gaussian Mixture Model (GMM) provides the most accurate phase detection. This result enhances the accuracy of flight phase identification and demonstrates the advantage of GMM in handling the intrinsic complexity of flight data.
Furthermore, our analysis points out areas of current research deficiency and proposes future directions for investigation, offering a roadmap for the development of machine learning methods in aviation that could improve operational effectiveness, safety and training programs.

Can urban retrofitting achieve a positive energy balance? A case Study of four European Positive Energy District

Abstract— Urban retrofitting has emerged as a key strategy in the transition towards sustainable cities, with Positive Energy Districts (PEDs) serving as a model for achieving energy-positive urban environments. This paper explores the potential for urban retrofitting to achieve a positive energy balance through a case study of four existing districts in European Municipalities: Settimo Torinese (Italy), Großschönau (Austria), Amsterdam (Netherlands), and Resita (Romania). The analysis leverages energy balance simulations, considering various retrofitting scenarios, including building insulation, photovoltaic (PV) installations, and the adoption of flexible grid usage. The findings indicate that while achieving a PED is challenging, it is attainable through a combination of aggressive retrofitting measures, renewable energy integration, and smart energy management. The study highlights the importance of context-specific strategies, as climatic and urban characteristics significantly influence the outcomes. It aims to add to the ongoing discourse on sustainable urban development by providing empirical insights into the pathways and challenges of achieving PEDs through urban retrofitting.
Keywords — Positive Energy Districts, PED, retrofitting, Climate-Neutral Districts

A Study on Reactive Power Control Using Variable Gain and Voltage Limiter for Grid-Forming Converters

In our previous study, an active power control method called variable gain control (VGC) has been proposed for grid-forming (GFM) converters. In the present paper, a reactive power control method for GFM converters to prevent an overcurrent and an alleviate a voltage oscillation under a three-line-to-ground (3LG) fault. In the proposed method, the magnitude of the voltage output of GFM converters is varied in accordance with the terminal voltage. In addition, a novel idea of variable gain and a voltage limiter is applied to avoid overcurrent. Numerical simulations demonstrate the effectiveness of the proposed method in suppressing overcurrent and voltage variation during 3LG faults.

Review of Mathematical Modelling and Interference Minimization Schemes for the Coexistence of 5G and Satellite Radio Access Networks

The aim of the study is to develop a suitable algorithm for interference minimizing in 5G and satellite communication networks coexistence employing Nakagami-m and Shadowed Rician models. Based on this aim, the following research contributes to:
1-Develop a suitable theoretical strategy that evaluates interference scenarios for co-existence between5G and satellite communication networks.
2-Develop an algorithm based on Nakagami-m and Shadowed Rician models for interference minimization in the co-existence between 5G and satellite communication networks.

Improving Generalization in Convolutional Neural Networks with a Dynamic Attention Layer

This paper introduces a novel Dynamic Attention Layer (DAL) that enhances the generalization capabilities of Convolutional Neural Networks in both in-distribution and out-of-distribution scenarios. By dynamically adjusting attention weights based on selected percentiles during training, DAL improves the network’s ability to capture both dominant and subtle features, resulting in better accuracy and robustness across diverse datasets. The study demonstrates DAL’s effectiveness through rigorous testing, showing it outperforms traditional attention mechanisms and data augmentation techniques, offering a valuable advancement in computer vision.

A Market, Economic, and Technical Analysis of a Community Solar Electricity Aggregator Approach for Local Governments to meet their Net Zero Emissions Energy Needs

This research can enhance the level of understanding of local governments in Western Australia on possible approaches to meet their net zero emissions targets.

Advancing Brain Tumor Detection via ViRCNN: A Fusion of Vision Transformers and Faster R-CNN

In the field of cancer diagnosis, especially detection of brain tumors, achieving highly accurate detection is very im- portant. Deep learning, with its remarkable capabilities in object detection, has emerged as a valuable tool for identifying brain tumors. We introduce a novel approach called ViRCNN that combines the strengths of Faster R-CNN and Vision Transformer (ViT), referred to as ViRCNN. This method enhances both the accuracy and efficiency of brain tumor detection in magnetic resonance image (MRI) images. To evaluate the effectiveness of ViRCNN, we employed the Br35H dataset, which includes 801 MRI images for training, validation, and testing. Our approach demonstrates significant improvements in the Mean Average Precision 50 (MAP50) and Recall metrics compared to previous methods. Notably, ViRCNN achieves a 0.9% improvement in the MAP50 score while maintaining a parameter count of only 19 million, substantially lower than the over 80 million parameters typical of state-of-the-art methods.

Assessing and Predicting Air Pollution in Asia: A Regional and Temporal Study (2018-2023)

This study provides a comprehensive temporal and geospatial analysis of PM2.5 levels across Asian countries from 2018 to 2023. Utilizing time series modeling (ARIMA) for predicting future pollution trends and evaluating multiple metrics for model performance, the research highlights significant regional variations in air quality and identifies key patterns in pollution trends. By categorizing countries into different pollution level clusters, the study presents a nuanced understanding of air quality dynamics, facilitating targeted policy recommendations for environmental management and public health interventions in Asia. This work contributes to the field by combining predictive modeling with spatial analysis to address a critical environmental issue.

Energy Consumption Forecasting Using Ensemble Machine Learning Models in Smart Grid

Energy Consumption Forecasting Using Ensemble Machine Learning Models in Smart GridAccurately predicting medical charges is crucial for healthcare providers, insurance companies, and policymakers to manage costs and allocate resources efficiently. This study conducts a comparative analysis of five machine learning algorithms—Linear Regression, Decision Tree Regressor, Random Forest Regressor, XGBoost Regressor, and Gradient Boosting Regressor—to evaluate their performance in predicting medical insurance charges. Utilizing a dataset of patient demographics, health indicators, and lifestyle factors, we identify the key variables that most significantly influence medical expenses. Our findings reveal that certain algorithms outperform others in predictive accuracy, with XGBoost Regressor showing the highest accuracy (R² = 0.94). Additionally, the study highlights the most critical factors contributing to medical charges, are Smoking status, BMI, and Age. The analysis of feature importance across different models provides valuable insights into the underlying drivers of healthcare costs. This research contributes to the growing body of literature on healthcare analytics by offering a dual focus on predictive modeling and variable importance. The results underscore the potential of machine learning to enhance decision-making in the healthcare industry, particularly in optimizing resource allocation and cost management.

Construction of a Regional Public Transportation Management Support System Using the Cloud

Public transportation is an indispensable part of residents’ daily lives, including commuting, shopping, and hospital visits. Our research group provides support activities for regional public transportation systems, mainly for community buses operated by local governments. The primary support services include the development of the General Transit Feed Specification (GTFS), bus location that displays the location of buses on a map, and the measurement of the number of passengers. We are supporting the DX of regional public transportation by constructing and providing an infrastructure system to realize these services. Until now, the infrastructure system was built on a server in a university laboratory to provide these services. However, the service was often interrupted due to power outages for legal inspections several times a year and network failures within the university, resulting in constant complaints from the regional public transportation operators they support. The conversion of the infrastructure system to the cloud solves these problems. A comparison of communication speeds showed that the on-premise environment was faster. However, by converting the infrastructure system to AWS, the security risk of converting their servers and the risk of power outages due to legal inspections can be reduced or eliminated, so these speed differences are considered acceptable.

Single-Stage PV-Grid Integrated Multilevel Inverter Driven Induction Motor Drive for Water Pumping

This paper presents an innovative single-stage grid-integrated solar photovoltaic (PV) system for water-pumping applications using an induction motor drive (IMD). The proposed system employs a seven-level diode-clamped multilevel inverter (MLI) to convert DC power from the PV array to AC power for the motor, eliminating the need for an intermediate DC-DC converter. A perturb and observe (P&O) algorithm is utilized for the PV array’s maximum power point tracking (MPPT). Direct torque control (DTC) with space vector modulation (SVM) provides precise speed regulation of the induction motor. The system enables bidirectional power flow between the PV array, motor load, and utility grid, optimizing energy utilization under varying irradiance and demand conditions. The proposed configuration exhibits enhanced efficiency and improved power quality compared to conventional two-stage topologies, offering a promising solution for grid-connected solar-powered water pumping systems.

Identifying Silent Mutations for the Introduction of Restriction Sites in Open Reading Frames

For many applications in molecular biology, restriction sites need to be engineered into an open reading frame (ORF), a part of the genetic material that codes for a protein. Importantly, silent mutations need to be performed because only these do not alter the amino acid sequence of the protein. However, finding such silent mutations is a very time-consuming process. We have developed a program which recognizes all silent mutations in an open reading frame (ORF) that each lead to a new restriction site. Our program uses python technologies comprising web crawlers and data analysis libraries to deduce the amino acid sequence coded by an ORF and convert the ORF nucleotide sequence into the amino acid single letter sequence. In doing so, reverse translation back into the nucleotide sequence allows the consideration of all possible nucleotide sequences coding for the same amino acid sequence, which are then compared with the restriction recognition sites of commercially available restriction enzymes, such as from New England Biolabs (e.g., https://www.neb.com/). This allows the identification of restriction sites that can be engineered via silent mutations within the provided DNA input sequences. The output is presented in a user-friendly tabular format that can be examined or downloaded (as a CSV file) for ongoing evaluations.

Next Level Chatbot: Expert Advisory Solution

The proposed research aims to enhance ticket creation efficiency by utilising a chatbot embedded with automation capabilities, allowing tickets to be completed without human intervention. This will significantly improve overall processing time and workflow from both a customer service and technical efficiency perspective.

North Atlantic Offshore Wind Characteristics: Modeling and Comparison with Field Measurements and Industry Standards

Wind characteristics are critical to offshore wind resource development. The power output from a wind turbine is very sensitive to the local wind speed. Wind speed measurement is often limited to surface area close to Lidar buoys or meteorological stations and up to 200m due to the range of remote sensing devices. On the other hand, wind fields from ground level and up to 20000m above ground level can be simulated using Weather Research & Forecasting (WRF) model. In this study, WRF simulations are performed for the North Atlantic offshore waters to obtain wind speed time and spatial properties. Statistics of wind speeds for selected sites are derived and validated with field measurements. Wind vertical profiles are compared with ISO and IEC standards, and a power law profile is further derived to find the best fit. It is also demonstrated that the WRF model is reliable to forecast wind data, optimize prediction and improve reliability for coastal and offshore energy development. The wind modeling and characterizing can be extended to global regions to identify prospects with the most renewable energy potentials.

Enhancing Routing Efficiency and Performance in Mobile Ad-Hoc Networks Using Deep Learning Techniques

Abstract— MANET stands for Mobile Ad-hoc Network also
called wireless Ad-hoc Network or Ad-hoc Wireless Network. It is
a decentralized wireless network consisting of mobile devices
(nodes) that communicate with each other without relying on a
fixed infrastructure. MANET forms a highly dynamic
autonomous topology with the presence of one or multiple
different transceivers between modes. MANETs consist of a peer
to-peer, self-configuring and self-healing modes. Mobile Ad-hoc
Network has wide range of applications such as military and
defense operations, healthcare, sensor networks, wireless sensor
networks, Internet of Things (IoT) etc. In order to enhance the
routing efficiency and performance in Mobile Ad-Hoc Networks
(MANETs) this paper proposing different deep learning
techniques.
Keywords— Mobile Ad-Hoc Networks, Wireless Network
Topology, Wireless Communication, Deep Learning.

Enhancing Routing Efficiency and Performance in Mobile Ad-Hoc Networks Using Deep Learning Techniques

Abstract— MANET stands for Mobile Ad-hoc Network also
called wireless Ad-hoc Network or Ad-hoc Wireless Network. It is
a decentralized wireless network consisting of mobile devices
(nodes) that communicate with each other without relying on a
fixed infrastructure. MANET forms a highly dynamic
autonomous topology with the presence of one or multiple
different transceivers between modes. MANETs consist of a peer
to-peer, self-configuring and self-healing modes. Mobile Ad-hoc
Network has wide range of applications such as military and
defense operations, healthcare, sensor networks, wireless sensor
networks, Internet of Things (IoT) etc. In order to enhance the
routing efficiency and performance in Mobile Ad-Hoc Networks
(MANETs) this paper proposing different deep learning
techniques.
Keywords— Mobile Ad-Hoc Networks, Wireless Network
Topology, Wireless Communication, Deep Learning.

Machine Learning based Multi-Variate MBB-User Growth Prediction and Worst-Cell Clustering in Cellular Network

The Paper herein, introduces the algorithm and the model of machine learning where Multiple variable linear regression model, support vector machine, K-Means clustering method is used to predict the mean user number depending of some other variables. This analysis will help a market operation and planning team of a telecom network, to design and optimize the network and achieve maximum users under a telecom network. The worst cells which we obtained from the clustering methods will help them to work with the worst cells and solve network issues or capacity issues to get more subscriber to improve their profit. This analysis will help to plan and design cluster by cluster which is also very important for a telecom operator. The final result correctly leads the company to predict the user number of a network, which definitely provides great commercial value and help to build a good and customer-centric mobile network. The predictive models and clustering algorithms provide actionable insights and recommendations for improv- ing network efficiency, QoS, and user satisfaction. Despite the promising results, the research faces several limitations and challenges, including data quality issues, model interpretability, and algorithm scalability. Future research directions may focus on addressing these challenges and exploring new techniques for enhancing predictive accuracy, model explain ability, and computational efficiency

Robust Cascade PID-based Controller Design for Brushless DC Motor using Antlion Optimization Algorithm

Brushless DC (BLDC) motors are widely utilized in various fields, including high-speed drives, artificial heart pumps, and electric vehicles, due to their superior torque, compact size, and enhanced efficiency. However, it is very difficult to obtain satisfactory control
performance for BLDC motors using conventional PID
controller, because of the difficulty in tuning the proper PID parameters.
In this paper, an optimal cascade PID controller is designed for
controlling the BLDC motor. The proposed cascade controller consist of an inner loop and an outer loop, each responsible for different aspects of the control process. The inner loop handles fast dynamics, namely current control, while the outer loop deals with slower dynamics, as speed control. This separation allows each loop to be optimized individually, resulting in improved overall system performance and stability. By quickly responding to disturbances in the inner loop, cascade controllers can effectively overcome oscillations and enhance the stability of the motor. Additionally, cascade controllers can better handle non-linearities and parameter variations, leading to more accurate and reliable control of the output. However, tuning PID controller gains is crucial for achieving optimal performance and stability in control systems, and metaheuristic algorithms offer significant benefits by efficiently searching for the best gain values, even in complex and high-dimensional parameter spaces. The proposed cascade PID controller’s gains are optimized using the Antlion Optimization (ALO) algorithm, a modern metaheuristic algorithm known for its effectiveness in constrained problems and diverse search spaces. This optimization enhances the controller’s robustness against disturbances, particularly supply voltage variations. To validate the system’s performance, Hardware-in-Loop (HIL) Typhon technology is employed, allowing real-time testing of the BDCM and controller under various conditions. This ensures the reliability and effectiveness of the system before actual implementation.

Creating Financial Management Prowess with AI-enabled Enterprise Systems

Insights from large and medium firms in this paper provide an understanding on how manufacturing firms can enhance financial management processes using artificial intelligence enabled enterprise systems. These results highlight the impact of these systems in developing financial management prowess and contribute valuable knowledge to both industry practitioners and academia.

Can AI Tell More than the Available Abstracts?

Predicting certain information from abstracts only via AI is a bold application of AI. The success of this project can significantly help biotech companies to market their products to potential customers. It can also broaden the application spectrum of AI.

A Low-Cost Deep Reinforced Hybrid Supercapacitor (DRHS) with Integrated Massless Energy Storage System (IMESS): Design and Development
Platypus Detection through Deep Learning

The research advances automated platypus detection using state-of-the-art models, achieving high precision and efficiency, significantly reducing manual classification in ecological studies.

An Integrated Blockchain-based Digital Twin Platform for Safety and Security in High-Risk Industries

1. The paper identifies and outlines the security challenges
faced by high-risk industries, particularly in sectors such
as energy, healthcare, transportation, and manufacturing,
where human lives are at stake.
2. It proposes an innovative solution that integrates digital
twin technology and blockchain to address the identified
security challenges. Digital twin technology enables
real-time virtual representations of physical assets, while
blockchain ensures data integrity and immutability.

Adaptable Wireless Power Transfer for Assistive Mobility Devices- A Review

This paper gives an overview on wireless power transfer (WPT) for assistive mobility devices (AMD). Shows how WPT can significantly contribute in enhancement of AMD for better inclusivity of persons with mobility challenges in the society as a whole.

Retrofitting Legacy CNC Machines with a Focus on Energy Consumption

We are outlining the detailed process of retrofitting a legacy CNC machine to add connectivity and integrate it into the Industry 4.0 framework. This work offers significant benefits, especially for small and medium-sized companies. The outcome has the potential to enhance sustainability in industrial processes by focusing on the machine’s energy consumption.

A Digital Twin Framework in Manufacturing Systems with a Focus on Energy Consumption

We have developed a framework for integrating digital twins into older CNC machines, focusing on monitoring their electrical energy consumption. The digital twin’s virtual component utilizes discrete event simulation software in a digital manufacturing environment. Our goal with this proposal is to promote greater sustainability in manufacturing systems.

Human Activity Recognition from Biometrics Data using Kolmogorov-Arnold Network

Human Activity Recognition (HAR) is a feature
of an automated system that recognizes human actions. Since
most people these days are health-conscious, people use their
smartphones or smartwatches to track their daily activities. This
helps them organize their schedules and lifestyles more effectively.
Recent advancements in Deep Learning (DL) performance have
mitigated certain issues related to HAR. Consequently, DL methods
are essential for improved competence and precision. This
paper provides a comparative study that utilizes state-of-the-art
Kolmogorov-Arnold Network (KAN) and Multi-layer Perceptron
(MLP) to classify human activities using biometrics data. The
Biometrics dataset, which includes 18 classes representing a
variety of activities, is used for HAR. For optimal outcomes, the
suggested algorithm is trained and tested using the TensorFlow
structure and a hyperparameter tuning technique. The outcomes
show that the KAN algorithm performs quite well in identifying
human activity with an accuracy of 72.64% and a loss rate of
0.9136. The experiment’s findings suggested that the KAN model
performs more effectively and accurately for human activity
identification.

Spatial Characteristics of CA: A Narrative into San Francisco

This study delves into the spatial characteristics of housing prices within California, with a specific focus on San Francisco. We explore various models to predict housing prices based on different feature sets, including coordinate and non-coordinate attributes. Through extensive analysis, we find that incorporating geographic coordinates significantly enhances the predictive accuracy of housing prices. Utilizing a neural network model, we achieve nearly 100% accuracy in predicting the quartile of median housing prices, underscoring the complexity and influence of location-specific features. Finally, we utilized i-SLFN algorithm for developing a precise price prediction model and describing communities characteristics.

Analyzing 5G Network Performance Using Interactive Gaming and Video Streaming Applications

The significance of the paper is to study how 5G networks respond and adapt to different application conditions.

How Much Data Do We Really Need for Question-Answering Benchmarks? A Study Based on SQuAD v2.0

This research investigates how many
question-answer pairs are required for fine-tuning
language models on question-answering (QA) tasks. We
fine-tuned nine different language models on subsets of
the SQuAD v2.0 dataset by Rajpurkar et al. (2018)
and measured the threshold at which the marginal
benefit of additional question-answer pairs diminishes.
We show that most fine-tuned language models for QA
often do not require the full SQuAD v2.0 dataset with
130,319 training samples to perform well; 78,191 samples
(60%) are in most cases enough to achieve near-peak
performance. Thus, smaller datasets may suffice to
fine-tune QA models. Competitive performance on
smaller datasets enables less resource-intensive training
of models, makes QA tasks more accessible without
requiring powerful hardware, and helps curators of
datasets to decide how much data to collect for a given
task.

Identification of the Most Frequently Asked Questions in Financial Analyst Reports to Automate Equity Research Using Llama 3 and GPT-4

This is the first systematic analysis of the structure of equity research reports (ERRs). The paper uncovers the automation potential of ERRs, and validates their automation potential with the state-of-the-art language models GPT-4 and Llama-3-70B.

The study examines 72 ERRs, categorizing 4,964 sentences into 169 unique question archetypes across five main categories: Financials, Company, Product, Stock, and Market. The researchers classify each question based on its potential for automation, distinguishing between text-extractable, database-extractable, and non-extractable information.

Key findings include:

– 78.7% of the questions in ERRs are potentially automatable.
– 48.2% of questions are text-extractable (suited for processing by large language models).
– 30.5% of questions are database-extractable.
– Only 21.3% of questions require human judgment to answer.

The study validates these findings using two large language models: Llama-3-70B and GPT-4-turbo-2024-04-09. The results show that these models can extract relevant information from annual reports for a significant portion of the questions, with potential for even higher accuracy when used in combination.

An Interpretable Transformer-Based Approach to Classify Malaria From Blood Cell Images

Malaria is a disease that can be fatal, and it is spread through the bite of the female Anopheles mosquito. The life of the sufferer is put in jeopardy as a result of the presence of numerous plasmodium parasites, which spread throughout their blood cells. Malaria can potentially be fatal if it is not treated within the first few stages of the disease. A well-known method for diagnosing malaria, microscopy involves taking blood samples from the patient, calculating the number of parasites, and counting the victim’s red blood cells. Nevertheless, the procedure of microscopy takes a lot of time, and, in certain circumstances, it can produce an incorrect result. When compared to the more conventional approach of microscopic examination, the recent successes of deep learning (DL) in the field of medical diagnosis make it quite conceivable to reduce the expenses associated with the diagnosis while simultaneously improving overall detection accuracy. This study proposes a transformer-based DL technique for diagnosing the malaria parasite using blood cell images. An explainable AI technique called Grad-CAM was applied in order to determine which aspects of an image the proposed model paid significantly more attention to in comparison to the other aspects of the image through saliency mapping. This was done in order to demonstrate the usefulness of the models. According to the findings of this research, the performance of the vision transformer and the VGG16 are identical. Both models have reached an accuracy score of approximately 96%, which is very impressive.

ResiPlant-5: A CNN model for disease detection in citrus fruits and leaves

This paper introduces ResiPlant-5, a 5-layer con-
volutional neural network (CNN) designed for precise plant
disease diagnosis. Sequential CNN model suffer form vanishing
gradient problem. To overcome vanishing gradients and im-
prove deep model learning, the design uses skip connections,
inspired by Residual Networks (ResNets). Skip connections
provide connections between non-adjacent layers, improving
gradient propagation and feature retention. This approach lets
the model maintain important properties from previous layers
while training deeper networks without sacrificing speed. Using
deep learning and residual connections, ResiPlant-5 successfully
addresses difficult image classification challenges, hence making
it feasible to identify the disease in the plants. The model has been
trained and tested using two publicly available datasets. The first
dataset is the citrus dataset, which contains images of citrus leaves
and citrus fruit. The second dataset is the sweet orange dataset.
The results indicate that the proposed model demonstrates an
approximate increase in accuracy of 2%, 6%, and 8% on the
Sweet Orange, citrus leaves, and citrus fruit datasets, respectively,
compared to the VGG16, VGG19, and ResNet50 models

Power System Faults Analysis, Detection, and Localization in Underground Distribution and Transmission Networks by Deploying AI-based Matlab Model/Simulink

The study integrates advanced AI algorithms into MATLAB/Simulink to enhance the accuracy and efficiency of fault detection and localization. It specifically addresses the unique challenges of underground distribution and transmission networks, which are often harder to monitor and diagnose than overhead systems. The research utilizes MATLAB/Simulink for detailed simulations, enabling robust testing and validation of fault detection and localization methods under various conditions.

PMU-Based Short Circuit Capacity Estimation using System and Load Impedance Variation Ratio

This paper presents a new method for the estimation of short-circuit capacity using PMU (Phasor Measurement Units) measurements of voltage and current phasors. The proposed method has two functions: the improvement of the estimation accuracy by selecting only PMU measurements which are suitable for the estimation, and the determination of whether the estimation values are erroneous or not. The proposed method focuses on variations in system side Thevenin equivalent (TE) impedance and load impedance, which affect the estimation accuracy. First, it analyses how the changes in each component of TE or load impedance affect the short-circuit estimation. Based on these results, the proposed method is developed. Then, the proposed method is validated by numerical simulations, and it is confirmed that the proposed method can properly distinguish between the correct and incorrect estimates and obtain more accurate estimation values of short-circuit capacity than other existing methods.

Detection of wood cross-section regions with GrowCut for measurement of wood diameter grade

Proposal of a Method for Detecting the Cross-sectional Region and Measuring the Minimum Diameter of a Single Timber Cross-section Image and Verification of the Detection Accuracy

Improvement in detection accuracy was confirmed by modifying the confidence level in GrowCut.

The effectiveness of forward-seeded pixels in reducing the influence of false positives in seeded pixels was confirmed.

The effectiveness of the RANSAC method for removing false detection contour points was confirmed.

Elevating 5G Applications Performance: Harnessing Beamforming with MIMO Antennas

This paper centers on the significance of using beamforming over Multiple Input Multiple Output (MIMO) at the physical layer level within a Radio Access Network (RAN) for 5G network performance.

Semantic Segmentation of Food through Deep Learning: A Case Study

In this paper, we explore food image semantic segmentation at different levels. We identify two datasets that can be used for training and evaluating models. We also assess the performance of four semantic segmentation models for food segmentation tasks. Our results show that FCN and SegFormer achieve the best overall accuracy at 85.9% and 84.1% when applied to the UECFOODPIXCOMPLETE dataset. The study aims to offer valuable insights and guide future developments in dietary assessment tools that can underpin health management applications.

Optimising Efficiency: Leveraging Multi‑Criteria Decision-Making for Field Instrument Selection in Process Plants

This study contributes significantly to the field of process plant instrumentation by:
Addressing Selection Challenges: Provides a structured approach to overcome the difficulties in selecting field instruments and sensors, ensuring adherence to specified conditions and maximizing operational efficiency.
Developing MCDM Method: Introduces a comprehensive Multi-Criteria Decision Making (MCDM) method tailored specifically for instrument selection in process plants, serving as a valuable resource for practitioners.
Automated Tool Creation: Develop an automated tool using Visual Basic for Applications (VBA) and built-in Excel formulas to facilitate the implementation of various MCDM techniques, streamlining the selection process.
Exploring Alternative Methodologies: Investigates the use of Decision Trees and Machine Learning for instrument selection, providing insights into their efficacy and potential for optimization, thereby broadening the scope of available methodologies.

A Phased Training Method for Stabilizing the Training Process of Power Grid Voltage Control Agents with Deep Reinforcement Learning

In recent years, renewable energy sources such as photovoltaic power generation system (PV) have been rapidly integrated into many power grids around the world. The higher penetration of renewable energy resources has made more difficult to maintain proper voltage using conventional method of Load Ratio Transformer (LRT) tap-changing in view of rapid generation fluctuation caused by weather condition change. To solve this problem, the reactive power control with power conditioning system (PCS) of PV can be used as a voltage regulation resource. Recent works have developed multi-timescale voltage control with short-term control by PCS and long-term control by LRT using deep reinforcement learning (DRL).
Most of these methods achieve coordination between agents in different control cycles by reward calculation. However, the training becomes unstable due to the improper management of change of each agent’s strategy, and it may not be possible to control voltage of power grid.
In this paper, the authors have proposed a phased training method to improve the stability of the training process for each agent that performs either LRT control or PCS control in power grid voltage control with DRL. The effectiveness of the proposed method is verified by numerical simulations using a power grid model with large PVs.

Creating Competitive Game Situation Using Cognitive Load

This study identifies effective interventions that can create competitive situations in a memory-based card game. The competitive condition entertains people of all ages in the game. A competitive game situation would create a flow state in players. This study focuses on Pelmanism as a memory-based game. The study identifies gimmicks to interfere with a player superior to the others with the player unnoticed to create a competitive situation in a memory game. Experimental results show that interference with higher cognitive load has a greater effect on the outcome of the game. In the experiment, the interference requiring high cognitive load degrades the correct answer rate from 33.2% to 24.0%. The results of the random forest analysis indicate the importance of influencing the working memory in the memory game. It means the cognitive load on the working memory can lead to game situations that all people enjoy.

Creating Competitive Game Situation Using Cognitive Load

This study identifies effective interventions that can create competitive situations in a memory-based card game. The competitive condition entertains people of all ages in the game. A competitive game situation would create a flow state in players. This study focuses on Pelmanism as a memory-based game. The study identifies gimmicks to interfere with a player superior to the others with the player unnoticed to create a competitive situation in a memory game. Experimental results show that interference with higher cognitive load has a greater effect on the outcome of the game. In the experiment, the interference requiring high cognitive load degrades the correct answer rate from 33.2% to 24.0%. The results of the random forest analysis indicate the importance of influencing the working memory in the memory game. It means the cognitive load on the working memory can lead to game situations that all people enjoy.

Semantic segmentation of block-divided images – Consideration on evaluation method –

Drones having high resolution cameras and sensors have become more available and cheaper. We use drone-captured images taken directly above the plantation to research how well the types and areas of fruit trees, as well as other features, can be recognized. Wide-area images are synthesized from numerous locally captured images taken by the drone and are then divided into blocks (image blocks) with a certain amount of overlap to increase the number of blocks available for training. In this paper, semantic segmentation is applied to these blocks, and their classification performance is evaluated. In these evaluations, it is clarified that the overlaps in the blocks make it difficult to properly separate the training data for training the semantic segmentation network from the test data for performance evaluation. To address these issues, data augmentation is applied to the test data, and the evaluation results are presented.

Unveiling the Dynamics of Customer Shopping Trends Using Machine Learning Algorithm: A Comprehensive Analysis of Demographics, Purchase Behavior, and Payment Preferences

The present paper deeply investigates consumer shopping behaviour guided by the methods of machine learning. Having used the assessment of demographic parameters, purchase behaviour, and payment tendencies, this research comes up with deep-rooted consumer behaviour trends diversified across the various categories of consumer segments. The research includes an extensive database where advanced Artificial Intelligence technologies like clustering, classification, and forecasting are used to extract invaluable intelligence. The results underline certain differences in consumption patterns of clients belonging to the same demographic group which include people of different ages, genders, income levels, and geographical regions. The inclusion of machine learning technologies allows for gaining a supply of information about consumer behaviour that helps to make decisions wisely having sustainable development in the complex and competitive retail environment.

Converting Remote Islanded Communities into a 100% Renewable Energy Based Grid Incorporating Solar Photovoltaic and Battery Energy Storage

Requirement for 100% energy from renewable resources in remote power systems with high levels of irradiation resource available. Also, not totally reliant on variable renewable energy sources alone for all energy supply.

The AI Pentad, the CHARME2D Model, and an Assessment of Current-State AI Regulation

The contributions of this article are threefold: 1) we first introduce the AI Pentad to better understand and identify regulatory intervention points within AI’s core components, 2) we present the CHARME$^{2}$D model, a universal framework that can help frame, construct, and evaluate legislative efforts, 3) we conduct a broad assessment of the AI regulatory progress of selected countries and regions against the CHARME$^{2}$D model to highlight strengths, weaknesses, and gaps. This comparative evaluation offers insights for future legislative work in the AI domain.

Estimation of Likelihood for Prosocial Behavior from Physiological Responses during Listening to Music

The study proposes a method to measure the likelihood for target people to take prosocial behavior using their objective physiological data as well as Prosocial behavior means attempts to help others without expecting any external reward. Conventional methods use questionnaires to identify people with a strong likelihood to engage in prosocial behavior. The conventional methods impose an undesirable burden on them because the questionnaires on prosocial behavior are unusual to them. To avoid the unusuality, the proposed method makes people listen to heart-breaking music, recording a time series of their electrodermal activities (EDA), which are physiological responses expressing their emotional changes. An experiment turns out people likely to take prosocial people would show emotional responses to both acoustics and lyrics. An experiment turns out people likely to take prosocial people would show emotional responses to both acoustics and lyrics.

Three-Way Task Scheduling Algorithm for Cloud Computing

Cloud task scheduling is a crucial aspect of a cloud computing system, and its scheduling technique directly influences cloud platform resource usage and user service quality. This study presents a cloud task scheduling optimization algorithm, known as CTSA-3WD, which aims to address the issues of load imbalance, low resource utilization, and lengthy job completion time. In the suggested approach, the execution duration of cloud jobs and the actual computing resources situation restrict the task set’s light-load and heavy-load functions. The algorithm is based on the fundamental idea of three-way decision-making, and separates the work set into three pieces according to the percentage of the two jobs inside it. The system focuses on three distinct task sets and determines an optimal scheduling strategy by employing the Max-Min algorithm for the task set containing a significant proportion of light-load jobs, the Min-Min algorithm for the task set with a substantial percentage of heavy-load studies, and a combination of the Min-Min and Max-Min algorithms for the task set that includes both light and heavy load tasks. Essential resources within the designated nodes are rearranged, and the task that is most suitable for the underutilized resources is assigned to them in order to achieve the objective of reducing the overall time required for completion. The experimental results conducted on the CloudSim simulation platform demonstrate that the CTSA-3WD algorithm, when compared to Min-Min, Max-Min, and selective scheduling algorithms, effectively enhances overall resource utilization, user service quality, and resource efficiency. Additionally, it enables improved load balancing across the entire system.

Steganalysis For Still Images With LSB Steganography Using Machine Learning Algorithms

This paper investigated the application of machine learning for steganalysis using a feature-based dataset extracted from still images with the Least Significant Bit (LSB) steganography. We evaluated several models and found that Artificial Neural Networks (ANNs) achieve the highest classification accuracy within practical training times. The accuracy, however, is limited to 93% due to constraints within the dataset. To overcome this barrier, more comprehensive datasets and/or models should be examined in future.

Fly Ash as Sustainable Modifier, Using RSM as Modelling and Optimization Tool

This study uses Response Surface Methodology (RSM) to investigate the effect of using fly Ash type F as partial cement replacement on the compressive strength in concrete pavement. The response surface Methodology (RSM) is increasingly utilized in concrete mix design, as it offers a more effective approach to analyzing and optimizing experimental responses. RSM outperforms traditional experimental design methods in various ways, such as reducing the number of required tests, thus lowering test costs, and identifying optimal input variables based on test results. It can construct a scientific mathematical model and offer insights into the impact of individual factors and factor interactions on test results within the specified numerical boundary.
Additionally, a three-dimensional response surface is created to illustrate the connection between preparation parameters and the response index, allowing for a clearer understanding of the relationship between each factor and the response value. Accordingly, The researcher has utilized the RSM method to assess the influence of various factors on concrete performance
The study also aims to optimize the fly ash percentage in the concrete mix design. To evaluate the impact, the research methodology utilized mathematical modeling and methodical experimentation. The experiment considered various variables, including the amount of fly ash and the length of the curing process. The experiment concluded that the ideal fly ash concrete included a fly ash concentration of 15% and a requirement for an 90 day curing duration to produce a peak compressive strength of 52 MPa, and according to the RSM optimization the fly ash concentration is 14.28% , considering 90 day curing duration achive a peak compressive strength of 51.28 MPa. These results highlight fly ash’s ability to improve concrete performance. These kinds of developments are essential for infrastructure projects such as airports, roads, and infrastructure, where long-term viability and environmental effects are major priorities, It also advances environmentally friendly construction methods by optimizing fly ash concrete mixtures using RSM modelling tool. It offers insightful information on how to maximize concrete strength while reducing environmental impact and points the way for the next improvements in concrete pavement engineering.
Keywords—Optimization, Response Surface Methodology, Concrete pavement, Compressive strength, Airport, Curing time.

Designing and Evaluating an Innovative Text Analytics Solution for Online Retailers’ Operational Decision Support

This research contributes to the field of text analytics by providing a structured methodology for developing and evaluating solutions that can transform customer feedback into actionable insights. The findings highlight the potential of such solutions to significantly improve decision-making processes and strategic planning in online retail, ultimately leading to enhanced customer satisfaction and business performance.

Designing and Evaluating an Innovative Text Analytics Solution for Online Retailers’ Operational Decision Support

This research contributes to the field of text analytics by providing a structured methodology for developing and evaluating solutions that can transform customer feedback into actionable insights. The findings highlight the potential of such solutions to significantly improve decision-making processes and strategic planning in online retail, ultimately leading to enhanced customer satisfaction and business performance.

Feature enhancement and matching algorithms for material ablation measurement in high temperature wind tunnels

Aimed at the special requirements for dynamic measurement of material ablation inside high-temperature wind tunnels, a binocular stereo vision system based on straight slider rail laser projection and high-speed camera capture is designed. A feature enhancement method for ablation measurement objects in high-temperature and high-enthalpy environments is proposed, and a mathematical expression formula based on multi-line laser feature enhancement description and extraction of the light strip centerline is derived for adaptive rapid feature matching. This formula takes into account the grayscale centroid, camera frame rate, and the correlation between line laser scanning ranges, effectively reducing search complexity and dependence on high-frame-rate cameras. Experimental results show that the system can complete a 200mm scan within 1 second at a distance of 1350mm. Experiments on planar objects and spherical convex surface platform under various conditions demonstrate that the system can control the total error within 0.5mm at the normal distribution confidence levels of 1σ, 2σ, and 3σ which proving the efficiency, accuracy, and high dynamic characteristics of this method for non-contact measurement of material erosion in high-temperature wind tunnel environments.

Predicting Electricity Market Price Using Machine Learning and Quantifying Dependency Beyond Renewable Energy

We offer (i) a detailed analysis on the impact of variables beyond renewable energy sources on electricity price, and (ii) a unified machine learning-based platform that integrates other diverse factors beyond renewable energy to improve electricity price forecasting.
Our machine learning models predict electricity price by quantifying the dependency on renewable energy and other important diverse factors under unified settings.

Numerical Methods for the Minimum Energy Among Three Dynamic Systems Governed by a Class of Weakly Singular Integro-Differential Equations

In this study, we presented numerical methods for determining the minimum energy state among three dynamic systems governed by a class of integro-differential equation with weakly singular kernels (Abel-type). These equations were developed from a class of integro-differential equations originating from an aeroelasticity problem. By weighting energy criteria for the three systems, we intend to numerically reveal the most stable energy state for the systems with various initial conditions and tracking targets. A part of the numerical scheme is constructed by interchanging the differentiation and integration operations in the integro-differential equation. Promising numerical results are provided.

Functionalities of harvesting machines for industrial intercropping use cases

This paper contributes by first describing industrial types of intercropping harvests and second deriving necessary harvesting machine/robot functionalities from the types. These findings are important to design the needed machinery in order to realize industrial intercropping use cases.

Fine-Tuning Pre-trained model GPT for Educational Domain-Specific Corpus

Providing students with effective academic advising using low-energy consumption by fine-tuning pretrained model.

Enhancing Lifecycle Sustainability through Optimized Supplier Quality Management in Heating Manufacturing

This research shows that improving supplier quality management for Company A’s wall-hung boilers leads to notable sustainability gains, including a 20% reduction in material defects, a 25% decrease in carbon emissions, and a 30% increase in product durability. These findings highlight the critical role of supplier quality in enhancing product lifecycle sustainability.

Blockchain-Based Vaccination Certification System: Cross-Chain Analysis Using EVM Platforms and NFTs

The COVID-19 pandemic has highlighted the critical need for robust and efficient vaccination record-keeping systems. Traditional paper-based methods are fraught with issues such as inefficiency, susceptibility to fraud, and interoperability challenges across different regions. This paper explores the potential of integrating blockchain technology, Non-Fungible Tokens (NFTs), and smart contracts to develop a secure, transparent, and universally recognized system for managing pediatric vaccination records. By leveraging the decentralized and immutable nature of blockchain, each vaccination record can be uniquely represented as a tamper-proof digital certificate. Smart contracts are employed to automate various processes within the vaccination system, ensuring data accuracy and integrity. This study presents a theoretical framework and a proof-of-concept implementation, demonstrating the adaptability of the proposed system across multiple EVM-supported blockchain platforms, including Binance Smart Chain, Polygon, Fantom, and Celo. The system aims to enhance the security, integrity, and accessibility of vaccine records, providing a scalable solution for pediatric healthcare.

A Mobile App for OpenStack-based Clouds

This is the first mobile application for managing large scale Cloud infrastructures based on the openStack technology. This underpins the national Cloud of Australia.

Application of Artificial Intelligence to Diagnose Neurological Disorders in a Wearable EEG Device

Our system integrates AI with EEG applications to diagnose neurological disorders. It is able to classify the specified mental health disorder and wirelessly transmit the data obtained from the EEG device to a remote server.

A Comparative Analysis of Deep Learning Architectures for Efficient Brain Tumor Detection

This article studies the effectiveness of deep learning (DL) algorithms in detecting brain tumors, focusing on disorders such as “Glioma-Tumor,” “Meningioma-Tumor,” “Pituitary-Tumor,” and “No-Tumor.” Magnetic Resonance Imaging (MRI) is the primary tool for identifying brain tumors, and the paper proposes a convolutional neural network (CNN) architecture for efficient tumor detection. The study explores various CNN models, including DenseNet121, ResNet50V2, DenseNet201, EfficientNetB2, VGG16, and MobileNet, which enhance classification accuracy. The models demonstrate high precision, recall, F1-score, sensitivity, and specificity in predicting brain tumor conditions.

An SVM-Based Identifying of Hate Speech and Abusive Language In Indonesia Tweets

Hate Speech and Abusive Language In Indonesia Twitter

CMS: A Consortium Management Solution for Decentralized Identifier Resolution

In summary, To address the challenges present in the unified resolution process of existing DIDs, the primary contribution of this paper is the proposal of a consortium management scheme for DID universal resolvers, employing blockchain smart contracts and IPFS decentralized storage technologies. This approach aims to decentralize management and enhance the robustness and security of the resolution process.

Blockchain Integration for Enhanced Traceability in Fijian Sugarcane Supply Chains

Blockchain provides a decentralized ledger for
documenting transactions and monitoring items across the supply
chain. Using a case study methodology, this paper investigates
blockchain adoption in Fijian sugarcane supply chains,
concentrating on the influence on traceability, quality control, and
stakeholder participation. The findings indicate that blockchain
integration improves product information authenticity, lowers
fraud, and boosts market access for Fijian sugarcane goods.
Despite achievements, difficulties such as technological complexity
and regulatory compliance remain. The study suggests increasing
blockchain integration, building industry standards, and
promoting policy frameworks. Future studies should look at
larger agricultural supply chain concerns and encourage
sustainable development in Fiji.

Design and Development of a Heart Attack Prediction Application Using Machine Learning

Heart attacks have been a leading cause of mortality around the globe,
contributing towards premature deaths. The burden of this disease has been significantly impacted due to the late deduction of the disease. Machine learning has been used to curb the late deduction by implementing models in technologies such as websites, software and mobile applications. However, the major challenge lies in developing technology which is widely used for practical applications. Most machine learning applications remain in the prototype stage and are not generally accepted in real-world practice. This paper defines the necessary means and methodologies to design and develop a machine learning-based application which is acceptable in the real world using the design science research methodology

DESIGN, SIMULATION, AND INTEGRATION OF 5MWp FLOATING SOLAR PV WITH 760MW KAINJI HYDROELECTRIC POWER PLANT

This research work assists in analysing accurate data and modelling a suitable model for integrating floating solar PV with hydropower plants. As a result, greenhouse gas emissions are reduced, and the best or most effective/proper system configuration is achieved. It can also benefit stakeholders and investors in implementing a hybrid system design and development.
According to the National Renewable Energy Laboratory (NREL), the critical differences between ground-mounted photovoltaic solar plants and floating photovoltaic solar plants

STUDENT PERFORMANCE ANALYSIS IN HIGHER EDUCATION USING INTEGRATED APPROACH OF MACHINE LEARNING TECHNIQUES

ABSTRACT: Keeping track of early indications regarding students’ progress helps academics optimize their learning tactics and focus on varying educational practices to make the learning experience successful. Machine learning applications can help academics to predict the expected weaknesses in learning processes and as a result, they can proactively engage such students in better learning experiences. This paper examines the effectiveness of the integrated approach of machine learning (ML) techniques in predicting students’ academic progress. Predicting student accomplishment is crucial in matters of higher education, as well as machine learning, deep learning, and its connections to educational data. The proposed idea not only predicts student accomplishment but also makes it simpler for educators and administrators to monitor students so that they can provide assistance and incorporate the training for the best results. This study presents the view of students’ performance prediction models and explores several clustering and classification strategies that significantly enhance the accuracy of classification, particularly when a training dataset is accessible. Through the use of machine learning clustering and classifiers, such as Fuzzy C-Means, Stochastic Gradient Descent, Support Vector Machine, XGBoost, Gradient Boosting and K-Nearest Neighbors algorithms, we categorize instances as either indicative of a good or bad condition. As a result, our classification models demonstrate high accuracy in predicting students’ performance disorder outcomes.

An Automated Diagnosis of Diabetic Retinopathy Grading Using DenseNet169

The significant contribution of the rapid and accurate DR detection and evaluation method proposed in this article is: 1) a rapid grading method to solve DR detection problems, and 2) the proposed method can effectively and accurately assist in completing DR screening by helping to automate DR detection and evaluation. This verifies that the proposed method can be used for large-scale DR medical imaging screening, effectively assisting doctors in achieving efficient diagnosis.

MACHINE LEARNING MODEL FOR EARLY DETECTION OF ASTHMA SEVERITY IN ADULTS

At present, asthma diseases have emerged as a major health concern around the world. Identifying asthma in both children and adults as early as possible is crucial for providing timely medical intervention to control the progression of this chronic disease. However, creating an accurate predictive model for asthma in both age groups has proven challenging. Currently, a small specific sample size and low accuracy are attained by the models in the research. For the development of a predictive model that is useful for clinics, limited research was done to analyze a broad population of adults and children. However, significant data was generated by the tele-monitoring of the patients but inappropriate assessment was done to make early predictions on asthma in adults. This research focuses on utilizing the potential of telemonitoring data to construct machine-learning algorithms that help in predicting asthma before manifestation. The main goal of this article ability to understand the efficiency of machine learning (ML) algorithms in opposing asthma-related disease assault, with a focus on “Severity-None”, “Severity-Mild”, and “Severity-Moderate” diseases. This study presents the view of Asthma Disease Prediction (ADP) and explores several clustering and classification strategies that significantly enhance the accuracy of classification, particularly when a training dataset is accessible. Through the use of machine learning clustering and classifiers, such as K-Means methodology, Multi-Layer Perceptron (MLP), Decision Tree (DT), Stochastic Gradient Descent (SGD), and Naive Bayes (NB) algorithms. we categorize instances as either indicative of a good or bad condition. Consequently, our classification models demonstrate high accuracy in predicting asthma disease outcomes.

MACHINE LEARNING-BASED LIVER DISEASE PREDICTION: ENHANCING DIAGNOSIS AND PROGNOSIS

Our research addresses the pressing global issue of liver diseases by developing a robust Liver Disease Prediction (LDP) system using comprehensive patient datasets. We evaluate and compare multiple machine learning algorithms such as K-Means, Logistic Regression, Decision Trees, and Support Vector Machines to accurately classify chronic liver conditions. Through extensive data analysis and confusion matrix evaluations, we demonstrate significant improvements in prediction accuracy, providing reliable tools for early diagnosis and intervention. This innovative application of machine learning not only aids healthcare professionals in managing liver disorders effectively but also reduces diagnostic workload, thereby enhancing overall patient care and medical outcomes.

Fuzzy Logic-Enhanced Oral Health Assessment in Substance Abuse Rehabilitation: A Novel Approach

This study presents a new method for evaluating oral health in substance misuse rehabilitation by using fuzzy logic into the diagnostic procedure. Our methodology improves the assessment of complicated oral health issues commonly linked to substance misuse by utilizing fuzzy logic’s capability to manage uncertainty.

Leveraging Deep Learning and Machine Learning for Enhanced Dental Diagnosis: A Review of Artificial Intelligence in Identifying Substance Abuse Related Oral Health

We conduct a thorough analysis of the present state of research on AI-driven systems used to identify substance addiction by utilizing dental imaging and patient data. Through the process of synthesizing many studies, we are able to identify the strengths, limitations, and areas of knowledge that are lacking.

Implementation of Wavelet Transform Based Convolution Neural Network Method for Detecting Image Forgery

The traditional forgery methods are not able to detect forgery images due to latest development in software to edit original images. A highly sophisticated technique to be developed to detect image forgery of new images which is edited with high end modern editing tools. At present, image forgery detection is a one of the challenging task to detect whether an image is from authorized or unauthorized. In some criminal cases images are the major evidence to prove criminal by forensic departments. Forensic departments are to use new methods to detect image forgery which will be useful to investigate criminal cases further. In this paper, we proposed discrete wallet transform and CNN method to detect image forgery. First input images are preprocessed then apply CNN method to identify image forgery into either slicing or copy move. The performance of proposed method is compared with existing algorithm and our method shown better results.

Educational Knowledge-Based System for Traffic Accident Analysis in Residential Street for Transportation Engineering Students

This study is to construct a knowledge-based educational system tailored for transportation engineering students and trainee civil engineers focusing on residential street scenarios to solve potential issues related to traffic accidents. The system is designed to facilitate learning in managing and mitigating these problems. The paper outlines innovative systems’ developmental and evaluative phases, encompassing knowledge acquisition, representation, system building, and verification/validation processes.
The initial phase involves acquiring knowledge through a comprehensive literature review, followed by extracting expert insights through interviews and questionnaires. Subsequently, the gathered knowledge undergoes documentation, analysis, representation, and transformation into computer software using the Visual Basic programming language. The system was verified and validated by extensive testing, including unit testing, integration testing, and user satisfaction testing, which was performed using questionnaires.

Patterns In Twitter Use During a Disaster: Content Analysis of 2023 Türkiye-Syria Earthquake Tweets

We analyze more than 400,000 tweets posted between 6-21 February 2023, and explore different use cases of Twitter networking site aftermath of the quake series. We carry out descriptive analysis of the tweets distribution, and analysis on hashtag agenda setting property. Topic distribution both in hashtags and tweet content is investigated.

An effective metaheuristic algorithm for traveling salesman problem

The space net divides the solution space of combinatorial problems into multiple regions, and each region stores landscape information. This information can guide the algorithm to find the better solution compare with ACO.

Traffic Violations Generation: Data Augmentation of Video Generation Based on Diffusion Model

– A system generates irregular-behavior videos in a steady quality, which can be recognized by most of the object detecting systems;
– Unlike other video generation models, this system does not require high-quality inputs;
– This system is able to run with restricted computing and training resources.

Design and Implementation of a High Performance Network Function Virtualization Platform

This study applied parallel processing of incoming packets to reduce processing time. Experimental results show that proposed mechanisms enhanced network performance, allowing efficient resource use, reducing network latency, and ensuring stable packet transmission to fit service requirements.

DataPoll: A Tool Facilitating Cross-Domain Big Data Research

We present DataPoll, an “end-to-end” Big Data analysis tool designed to simplify the process and enhance accessibility for scientists across disciplines. DataPoll introduces innovative features and techniques for analyzing and interpreting digital data. Its capabilities and effectiveness are demonstrated through a case study on multi-source data from the Ukrainian-Russian conflict.

A Hybrid Approach: Machine Learning and Blockchain in Health Insurance Fraud Detection

This research introduces a system that integrates machine learning with blockchain technology, ensuring data transparency, security, and immutability while enhancing predictive accuracy. Demonstrated with real-world health insurance data, this hybrid approach significantly improves fraud detection accuracy and efficiency. Advanced machine learning algorithms provide insights into patterns and anomalies, enabling proactive fraud prevention. The solution is scalable and adaptable to other sectors prone to fraud. The use of Hyperledger blockchain ensures robust data integrity and security, addressing challenges related to data tampering and unauthorized access. These contributions collectively advance fraud detection and prevention in the health insurance industry.

An Effective Method for Classifying Japanese Honorific

Japanese Keigo known as honorific, is a way to reflect social status, intimacy, and the relationships among speakers, listeners, and other participants in a conversation. It is a very special and important language phenomenon that conveys respect and politeness based on the social status of the speaker and listener and their relationships. Unlike many other languages, Japanese has various forms of honorific expressions, and these honorific forms change depending on social group relations and occasion fields.

The hyperparameter tuning of a Multilayer Perceptron for agricultural decision classification in Gabon

This study randomly experiments different combinations of the multilayer perceptron’s hyperparameters, to find those that best improve our model’s performance.

Evaluating Lightweight Asymmetric Cryptography for Secure Communication in Internet of Drones

Unmanned aerial vehicles (UAVs) are being successfully used in a variety of applications, including agriculture, search and rescue operations, surveillance systems, and mission-critical services, thanks to some technological and practical advantages, such as high mobility, the ability to extend wireless coverage areas, or the capacity to reach locations inaccessible to humans. In contrast, attacks against drones, as opposed to traditional cyberattacks, typically happen as a result of serious design flaws and a lack of wireless security protection methods. The study examines lightweight asymmetric cryptographic algorithms for secure Internet of Drones (IoD) communication, addressing cybersecurity
challenges within this emerging technology. It evaluates RSA, ElGamal, DiffieHellman, and Elliptic Curve Cryptography (ECC), focusing on their suitability for IoD through comparative analysis on calculation time, memory usage, key size, and security. The goal is to contribute to developing robust, efficient, and secure communication protocols for IoD, promoting growth while mitigating risks. This research is pivotal for the advancement of IoD security, exploring the application of these cryptographic techniques to ensure secure, efficient operations within the IoD framework.

Efficiently Using Deep Learning to Distinguish Early-Stage Hepatocellular Carcinoma (HCC) from non-HCC Based on Multi-Phase CT and Image Enhancement

This study uses image enhancement methods to analyze liver nodule progression and radiological features in liver cancer development. A detection strategy has been developed from CT image characteristics for early identification of liver cancer. The study also explores how the size of nodules influences detection accuracy and classification between benign and malignant types, which is vital for refining detection algorithms and improving diagnostic precision.

DuoDistill: A Dual Knowledge Distillation Framework for Optimizing Top-K Recommender Systems

This work presents a novel knowledge distillation framework that utilizes multiple intermediated assistant models of varying sizes and architectures to facilitate knowledge transfer from a teacher (source) model to a student (target) model.