INTI Institutional Repository
Not a member yet
2178 research outputs found
Sort by
Predicting Parkinson’s Disease Using Machine Learning with Voice Parameters and Handwriting Images
No full textMost studies have failed to focus on geriatric diseases in the present era of quick advancement in medical science. Diseases like Parkinson’s display their symptoms at a later stage and make a complete recovery almost doubtful. Parkinson’s disease is a neurodegenerative disorder that affects movement and motor control systems. It is named after Dr. James Parkinson, the first person affected by this disease. Parkinson’s slowly worsens over time, leading to a variety of syndromes that can impact a person’s daily life activities. More than 95% of Parkinson’s Disease (PD) patients stated that they have exhibited voice impairment and micrographic disability. This model takes advantage of both advanced machine learning algorithms and modern image processing techniques, resulting in effective and efficient prediction PD. To further enhance the accuracy of the model, we have incorporated additional algorithms such as Random Forest and K-nearest Neighbour. Random forest classifier has a detection accuracy of 92%and sensitivity of 0.95%. The performance has been assessed with a reliable dataset from the University of California Irvine Machine Learning repository for voice parameters and a dataset from Kaggle for Handwriting images which includes wavy images and spiral images. Our proposed model has achieved the highest accuracy of 95% which outperformed the previous model or experiment on the same dataset
Web Engineering Methods in Building aWeb-Based School Academic Information System
No full textThe advancement of computer technology has paved the way for the development of computer-based information systems, significantly enhancing the efficiency and speed of data processing tasks. In educational settings, particularly in secondary vocational schools, there is a pressing need to transition from traditional manual data handling methods to more sophisticated, automated systems. This study focuses on the development of an academic web application aimed at improving the data processing quality in a secondary vocational school where data is currently managed manually using paper and pen.The existing manual method of processing academic data could be more efficient, prone to errors, and susceptible to data loss. This antiquated approach hinders the school's ability to manage academic information effectively, leading to delays and inaccuracies in data reporting and access. There is a clear necessity for a system that can streamline data processing, ensure data security, and provide easy access to information.The method employed to design this academic information system is web engineering. Web engineering combines engineering principles, management practices, and systematic approaches to create high-quality web-based applications. The development process includes several critical activities: formulation, planning, analysis, engineering, page generation, and testing. The system is implemented using PHP and MySQL, which were chosen for their robustness and reliability in web application development.The development and implementation of the academic information system have demonstrated significant improvements in data processing quality. The system allows for efficient, accurate, and secure handling of academic information. It facilitates quick accessto data from any location, enhancing the practicality of data management. Testing confirmed that all system functions met the predefined requirements, ensuring that the application operates as intended. The introduction of this web-based system has effectively addressed the deficiencies of the previous manual method, providing timely and accurate reporting. This improvement has enhanced the services provided to students, teachers, and principals, leveraging modern technology to achieve greater efficiency in educational administration
Q-switched Fiber Laser with Silver Nanoparticles-PVA in an Erbium-based Ring Configuration
No full textThis study examines the fabrication and experimental setup of a Q-switched fiber laser using a silver nanoparticle-polyvinyl alcohol (AgNPs-PVA) thin film as a saturable absorber. The thin film was produced by combining 5 mg of silver nanoparticles with 50 mg of polyvinyl alcohol in ionized water, followed by molding and drying. The laser showed a clear peak at 1560 nm, indicating precise output adjustment, with a slope efficiency of 9.69%. Pump power ranged from 49.06 to 81.37 mW, resulting in output power between 4.27 and 7.4 mW and pulse energy increasing from 100 nJ to 136.68 nJ. An optical-signal-to-noise ratio (OSNR) of 75.16 dB and a stable RF spectrum demonstrated strong laser stability. This work's competitive performance, ease of fabrication, and promising results suggest that the AgNPs-PVA thin film is a viable candidate for Q-switched fiber lasers, advancing laser technology
Voice-Assisted News App using Natural Language Processing
No full textIn recent times, the advancements in Artificial Intelligence (AI) and natural language processing (NLP) have enabled the development of novel voice-assisted apps. This study introduces a cutting-edge Voice-Assisted News App that utilizes the capabilities of Alan M, a prominent AI platform that revolutionizes the way people access news. This research endeavor attempts to utilize Alan Al's capabilities to offer a personalized news experience and enhance clarity through voice interactions. Users of the Voice-Assisted News software can effortlessly request and select their preferred news topics, and the computer will generate the most relevant news updates. Through integration with diverse news sources and the application of sophisticated algorithms, the program generates an individualized news stream for every user, ensuring they receive the latest and most relevant information. Users can efficiently access story audio descriptions. The app's speech interface allows users to navigate across different news categories. The NLP algorithms developed by Alan AI offer precise understanding of user demands by ensuring smooth interaction and an authentic conversational experience
The Impact of Copper on the Mechanical Properties of Cast Motorcycle Cylinder Blocks
No full textEnhancing the mechanical properties of cast products can be achieved by adding other elements. This research investigates the effect of copper (Cu) on the mechanical properties of cast motorcycle cylinder blocks, verified through microstructure analysis.Experiments were conducted by adding varying amounts of copper (Cu) at 5%, 10%, and 15%, with a casting temperature of 1010°C using a metal mold. Data collection included hardness testing, tensile strength testing, and microstructure analysis.The results showed that increasing the percentage of copper (Cu) in the cylinder block casting led to significant improvements in mechanical properties. The highest Vickers hardness was achieved with the addition of 15% Cu, measuring 229.23 HV. Similarly, the highest tensile strength was observed with the 15% Cu addition, reaching 104.218 MPa. Microstructure analysis revealed that the casting results predominantly consisted of α Al+θ (CuAl₂)+βSi phases, which were smaller compared to the α Al phas
Effective Bug Triage for Software Development and Maintenance
No full textSoftware businesses allocate about 45% of their budget to resolving issues. Bug triage is an essential step in the bug-fixing process that aims to effectively provide a developer with information about a new bug. This research focusses on the issue of data minimization in bug triage, which involves reducing and enhancing the quality of bug data. Utilize instance and feature selection techniques to simultaneously decrease the size of both the word and data dimensions related to bugs. The objective is to construct a prediction model for a novel bug data set by utilizing qualities from previous bug data sets. Additionally, we aim to assess the comparative significance of employing feature and instance selection in the sequence in which they are implemented. Empirically evaluate the effectiveness of data reduction by analyzing a total of 600,000 bug reports from two significant open-source projects, Mozilla and Eclipse. The findings indicate that our data reduction technique has the potential to effectively decrease the bulk of data while enhancing bug triage accuracy. Our study effort presents a methodology for utilizing data processing techniques to provide superior, sparsely populated bug data for the sake of software development and maintenance
Integrating HOG-Based Vehicle Detection with CNN-Based Lane Detection for Autonomous Driving
No full textThe advancement of autonomous driving systems hinges on accurate and reliable vehicle and lane
detection. This paper presents an integrated method to improve autonomous driving systems by
merging Histogram of Oriented Gradients (HOG)-based vehicle detection with Convolutional
Neural Network (CNN)-based lane detection. HOG effectively identifies vehicles by capturing
edge orientations and structural features, while CNNs excel in detecting intricate lane patterns
through deep learning. The combination of these techniques offers a robust solution for detecting
both vehicles and lanes, essential for autonomous navigation. Evaluated across a diverse dataset
featuring various driving conditions, the system's performance is measured using precision, recall,
F1 score (for vehicle detection), and accuracy (for lane detection). The results indicate significant
enhancements in detection capabilities, leading to improved situational awareness and safer
navigation. Future work will aim to refine the system further and tackle challenges in more
complex driving environments, marking this approach as a promising advancement in autonomous
driving technology
Optimizing Renewable Energy Integration in Weak Grids withUPQC Controller
No full textIntegrating renewable energy into weak grids poses significant challenges, including voltage instability, power quality deterioration, and limited capacity to handle variable generation. This paper proposes a methodology to improve the Power Quality in the weak grid with renewable energy penetration using additional Unified Power Quality Conditioner (UPQC) Controller. Bird Swarm Optimization algorithm is implemented in the conventional scheme to minimize the oscillations in the voltage signals.The proposedcontrol system evaluates the switching signals for the UPQC while accounting for variations in the grid's strength, wind speed, load currents, and dc link voltage dynamics. The efficiency of the suggested strategy for improving the PQ performance of weak grid-connected renewable energy sources in the presence of nonlinear loads was demonstrated by simulations conducted in MATLAB and experimental research
Enhancing Sustainability in Academic Guidance: Develop an AI-Driven Agent for Education 5.0
No full textThis study aims to develop a Multimodal Artificial Intelligence (AI) Agent named "Academic
Quick Guide", specifically designed for academic guidance in the Education 5.0 era. By
utilizing generative AI techniques and following Design Science Research Methodology
(DSRM) processes, the research seeks to create a user-centric AI Agent that streamlines
education management and improves academic advisory efficiency. Incorporating advanced
features such as Knowledge Databases and Prompt Engineering, this AI Agent is expected to
enhance user experience, facilitate decision-making, and improve academic outcomes
sustainably. Serving as a 24*7 self-service academic advisor, this AI Agent will be available
around the clock to support student' academic needs in a timely and effective manner.
Leveraging the Qwen Large Language Model (LLM) and the concept of "Model as a Service"
(MaaS), this AI Agent will promote sustainability in educational environments by optimizing
resource utilization, enriching learning experiences, and providing personalized academic
support
Convolutional Neural Network Model for Bone Fracture Detection and Classification in X-Ray Images
No full textBone fractures are one of the most common medical conditions worldwide. Proper and rapid
diagnosis of fractures is essential to ensure effective treatment and reduce the risk of further
complications. This study uses a Convolutional Neural Network (CNN) for fracture classification
on X-ray images, which aims for the clinical implementation of CNN models in supporting the
diagnostic process in the orthopedic field to minimize misdiagnosis due to human error. The
analysis results show that fracture classification using CNN has accuracy, precision, recall, and
F1-score reaching 99%, indicating highly accurate classification performance. This research aligns
with the 3rd SDG's goal of good health and well-being: to ensure a healthy life and support wellbeing.
The results of this research are expected to significantly contribute to the medical world,
especially in improving the accuracy and efficiency of fracture diagnosis and become a foundation
for developing more innovative diagnostic technologies to support more equitable and quality
health services globally