Journal of Computing Research and Innovation

Thyroid Insight: Navigating Disease Data Through Interactive Visualization with Prediction

Mon, 01 Sep 2025 00:00:00 +0000

The thyroid gland, located in the neck, plays a crucial role in regulating metabolism, growth, and energy through hormones such as thyroxine (T4) and triiodothyronine (T3). Disorders such as hypothyroidism, hyperthyroidism, and thyroid cancers are often linked to iodine deficiency and genetics. However, limited public awareness and delayed diagnosis can lead to severe health complications. Analysing thyroid disease data is challenging due to its complexity and unstructured nature, making advanced analytical techniques essential. This paper addresses these challenges by developing an interactive dashboard with predictive capabilities. The system integrates Big Data analytics and predictive modelling to improve understanding and support proactive management of thyroid health. It follows a structured methodology, including planning, analysis, design, development, and testing, using data from Kaggle and the UCI Machine Learning Repository. The dashboard employs Microsoft Power BI for visualizations and the Random Forest algorithm for predictive modelling. Evaluation using the Technology Acceptance Model (TAM) with 35 respondents produced encouraging results across dimensions such as Perceived Ease of Use (4.28), Perceived Usefulness (4.61), Attitude (4.54), and Intention to Use (4.50). User feedback highlighted the dashboard's intuitive design, clarity in presenting complex information, and potential to raise awareness about thyroid health. While the findings are based on a limited evaluation, results indicate that the system may contribute to improving public awareness, supporting early detection, and empowering users to make informed health decisions. With future improvements, such as real-time data integration and expanded datasets, the system could further enhance healthcare practices and public education regarding thyroid diseases, promoting proactive health management.

User Acceptance Testing of the ‘Furwish’ Location-Based Mobile Application: An Empirical Study on Pet Adoption and Surrender Services

Mon, 01 Sep 2025 00:00:00 +0000

The increasing number of stray animals demands an effective way to connect pet owners who intend to surrender their pet with potential pet adopters. The manual approach to pet surrender and adoption often faced many challenges, such as limited outreach, communication, and options for the adopters. Unadopted pets may lead to an increasing number of stray animals and cause disturbance in residential areas. In this paper, we present a more detailed discussion on the development and evaluation of a location-aware mobile application named FurWish. The mobile application allows users to post information about their pet for listings, location-aware pet searches, setting up appointments for adoption, and notifications. The development adapted a three-phase methodology from Waterfall Model which includes requirement identification, design and development, and evaluation. Requirement identification involved reviews of literature and existing mobile apps. In the design phase the user interface design is drafted using diagrams and sketches, and the development phase involved implementation using Android Studio and Firebase with the integration of Google Location Services for displaying maps. In the evaluation phase, a user acceptance test (UAT) is conducted to test four key aspects, which are Attitude Towards Using (ATT), Behavioural Intention to Use (BI), Perceived Usefulness (PU) and Perceived Ease of Use (PEU). 33 respondents were given 12 questions after they had explored and used the functions and features of the mobile application. Results from UAT show FurWish has achieved high scores over four key aspects, with the highest score achieved under PU. This shows the user's acceptance of its effectiveness in supporting the pet adoption and surrender process while supporting animal welfare.

CNN Integrated Mobile Application: Food Image Recognition for Recipe Generation

Mon, 01 Sep 2025 00:00:00 +0000

The rapidly advancing of the digital world has encouraged the use of mobile applications in almost every aspects of our everyday life. This includes transforming the way we obtain our meal, whether to order from food providers or simply cook for ourselves. The CNN Integrated Mobile Application for Food Recipe Generation is intended to improve users' culinary experiences by offering suggestions for recipes and intelligent ingredient recognition. This research explores the Convolutional Neural Networks (CNN) algorithm to tackle the problems of effective ingredient management and minimize food waste through smartphone application. Through the mobile application, the ingredient photos can be scanned by users or uploaded, after which the CNN model processes the images to precisely identify the ingredients. The application provides users with a wide range of meal alternatives that are customized to their available ingredients by retrieving relevant recipes from external databases such as the Spoonacular API based on the recognized ingredients. The research methodology consists of 3 main phases, which are Data Preprocessing, CNN Implementation and Performance Evaluation. In this research, the CNN algorithm has generated a good and acceptable performance with more than 96% accuracy. This research has shown how machine learning, mobile development, and user-centric design can be successfully combined to create a useful tool for contemporary culinary demands. The app encourages a move towards more sustainable and thoughtful eating habits by acting as an incentive for change at the community level. When communities embrace these ideas, the app plays a key role in tackling more significant social issues associated with food waste, supporting international initiatives that are detailed in the UN Sustainable Development Goals (SDGs) for a more sustainable and responsible society.

Evaluating User Acceptance of the UnityHub Web-based System in Higher Education: A Descriptive Analysis

Norazah Umar, Nurhafizah Ahmad, Jamal Othman, Rozita Kadar (Author) — Mon, 01 Sep 2025 00:00:00 +0000

The transition to hybrid learning in Malaysian higher education created a need for centralized platforms that enhance academic communication and class management. UnityHub was introduced as a web-based system to simplify group registration and strengthen student–lecturer interaction. The purpose of this study was to assess students’ acceptance of UnityHub using the Technology Acceptance Model (TAM) and to examine both the overall level of acceptance and the relative strengths and weaknesses of its dimensions. A total of 90 students from Universiti Teknologi MARA, Penang Branch participated in the survey, which used a TAM-based questionnaire comprising 24 items. Data were analyzed using descriptive statistics, including mean scores and standard deviations. The findings show that satisfaction scored highest (mean = 4.16), followed by attitude (3.80) and ease of use (3.78). Perceived usefulness (3.71) and intention to use (3.72) were moderately high, while self-efficacy was lowest (3.42). Overall, UnityHub achieved strong acceptance for usability and efficiency, though improvements in academic integration and user confidence are needed. The study is limited to a single department sample; future research should involve multiple faculties and compare UnityHub with other platforms to enhance generalizability.

Cyberpark: An-IoT based Automated Parking Management and Monitoring System using RFID and ESP32

Mon, 01 Sep 2025 00:00:00 +0000

The growing number of vehicles in urban and institutional areas has created significant challenges in parking space management. Traditional parking systems, which rely on manual processes, are often inefficient, time-consuming, and lack real-time information, leading to increased traffic congestion, fuel consumption, and user dissatisfaction. To address these issues, this paper presents Cyberpark, an IoT-based Automated Parking Management and Monitoring System that integrates smart technologies to improve parking operations, enhance user convenience, and increase infrastructure safety. Cyberpark utilizes RFID technology for secure and automated vehicle access, infrared (IR) sensors to detect parking slot occupancy, servo motors to control entry and exit gates, and a water sensor to detect flooding within the parking area. The system incorporates two microcontrollers: Arduino UNO for local control and ESP32 for wireless communication with the Adafruit IO cloud platform. Real-time data from all sensors is uploaded to the cloud, allowing for remote monitoring and status updates via an online dashboard. The development process involved circuit simulation using Proteus 8 Professional and functional testing through the Wokwi simulator. Hardware components were assembled on a custom PCB, and the software was developed using the Arduino IDE. Results showed effective system performance in automating gate operations, accurately monitoring parking slot status, and issuing timely alerts during flood conditions. The real-time dashboard allowed for seamless user interaction and remote oversight. Cyberpark demonstrates the feasibility of integrating low-cost, open-source hardware with cloud-based IoT platforms to create a scalable, user-friendly, and environmentally responsive smart parking solution. This system is suitable for deployment in campuses, shopping complexes, or residential areas. The project contributes to the development of smart city infrastructure by offering a reliable, efficient, and safe approach to modern parking management.

Enhancing Pineapple Cultivar Classification: A Framework for Image Quality, Feature Extraction, and Algorithmic Refinement

Mon, 01 Sep 2025 00:00:00 +0000

Accurate classification of pineapple cultivars is hindered by limitations in image acquisition, feature extraction, and classification algorithms. This study identifies these technical challenges and proposes methodological improvements to support reliable autonomous recognition systems. Key findings reveal deficiencies in current image datasets, leading to a proposed standardised acquisition protocol involving consistent lighting, optimal camera positioning, and suitable file formats. The HSV colour space is validated as more effective for extracting skin features, with its threshold values and post-processing steps significantly reducing classification errors. The proposed algorithmic refinements integrate chromatic and morphological attributes, particularly surface area and optimise logical operators to enhance accuracy. The study addresses two main research gaps: precise quantification of skin colour and the development of robust classification frameworks. Future work will emphasise empirical validation and the deployment of the YOLOv7 model for real-time, on-site assessment of fruit maturity in field conditions. These contributions hold strong implications for advancing precision agriculture and improving post-harvest processing.

Heart Failure Detection Using Scaled Conjugate Gradient Method and Naïve Bayes

Mon, 01 Sep 2025 00:00:00 +0000

Heart failure known as high mortality rates is a serious pathophysiological condition characterized and substantial long-term healthcare costs. Early detection is crucial, as the disease tends to progress without timely and appropriate intervention. This study aims to predict the risk of heart failure using structured clinical data and to leverage deep learning techniques to enhance the accuracy of risk assessment. The core objective is to demonstrate that early identification of heart failure indicators can significantly improve patient outcomes, potentially distinguishing between life and death. Recognizing these early warning signs provides a better opportunity for preventive care and timely treatment. To achieve this, two algorithms were employed: the Scaled Conjugate Gradient method within an Artificial Neural Network (ANN) framework, and the Naïve Bayes classifier. A Feed-Forward Neural Network (FFNN) was utilized as the primary classifier to detect the presence of heart failure. The neural network architecture used in this study consisted of 12 input neurons, 20 hidden layers, and a single output layer. The performance results revealed that the ANN achieved an accuracy of 86.7%, while the Naïve Bayes classifier reached an accuracy of 76.9%. Overall, the ANN demonstrated best performance in detecting heart failure, especially with a large number of hidden neurons, highlighting its potential as an effective diagnostic tool.

Determinants of Halal Food Purchasing Decisions Among Undergraduates at UiTM Tapah

Mon, 01 Sep 2025 00:00:00 +0000

This study explores the factors influencing halal food purchasing decisions among undergraduates at Universiti Teknologi MARA Perak Branch, Tapah Campus. With the rising demand for halal products among Muslim consumers, understanding these drivers is vital for business and policy makers. The study examines the roles of religiosity, knowledge, perception, awareness, attitude, and branding and promotion in shaping student’s purchasing decisions. A quantitative research design was employed, utilizing a structured questionnaire distributed to conveniently selected 104 undergraduates from various programs. The study focused on six determinants such as Religiosity, Knowledge, Perception, Awareness, Attitude, and Brand and Promotion Influence. Meanwhile, the dependent variable is Halal Food Purchasing Decision. Multiple Linear Regression analysis was conducted using IBM SPSS version 23. The results revealed that three determinants such as religiosity, awareness, and attitude significantly influence halal food purchasing decisions as p-value less than 0.05. Specifically, religiosity (B = 0.177) and attitude (B = 0.332) showed a positive effect , while awareness had a negative effect (B = - 0.114). This study provides valuable insights for food manufacturers, marketers, and university administrators aiming to align strategies with young Muslim consumer’s needs. It also highlights the importance of educational initiatives in enhancing halal knowledge and awareness. Future research could expand the scope by conducting comparative studies across different campuses or student demographics.

Exploring the Effect of Shape Parameters on Font Design Using Rational Quadratic Trigonometric Curves

Mon, 01 Sep 2025 00:00:00 +0000

This study investigates the impact of shape parameters on font contour design using rational quadratic trigonometric Bézier (RQTB) curves. By varying parameters (m, n) and weight (v), the research evaluates contour accuracy through pointwise deviation analysis and computational efficiency. The configuration (m = 0.5, n = -0.5, v = 1) achieved the lowest RMSE and CPU time, indicating optimal performance. An interactive tool was also developed to support real-time parameter adjustment. The results demonstrate that carefully tuned shape parameters enhance both visual fidelity and efficiency, making RQTB curves practical for typography and related design applications.

AI-Driven Forgery Detection in Offline Handwriting Signatures: Advances, Challenges, and the Role of Generative Adversarial Networks

Mon, 01 Sep 2025 00:00:00 +0000

Handwriting-based authentication continues to be a critical element in forensic analysis, particularly in the context of document fraud and signature forgery. Although deep learning (DL) techniques have shown promising results, there are still obstacles associated with the availability of limited datasets, the generalization of models, and their robustness. This review conducts a systematic examination of recent developments in DL methods for signature forgery detection. It employs the PRISMA protocol and retrieves literature from four well-established databases: Scopus, ACM Digital Library, Web of Science, and IEEE Xplore. Following a rigorous screening procedure, a total of 15 primary studies published between 2019 and 2025 were selected from an initial 115 records that were filtered by Computer Science subject area, English language, and original research articles. Five publicly accessible datasets: CEDAR, BHSig260, ICDAR 2011 SigComp, Kaggle signature verification dataset by RobinReni, and Kaggle handwritten signatures by Divyansh Rai were identified and analysed. The review indicates that Siamese networks dominate the DL architecture for signature forgery detection tasks, while alternative methods either employed fine-tuned pre-trained models (i.e., VGG16) or a hybrid of autoencoders and Convolutional Neural Networks (CNNs). An accuracy of 100% has been achieved through utilization of Siamese network leveraging the CEDAR dataset. This result is reasonable since CEDAR has the advantages of clean and balanced dataset. In response to the persisting limitations, this review emphasizes Generative Adversarial Networks (GANs) as the powerful data augmentation technique and a potential solution to enrich training datasets, simulate diverse forgery patterns, and enhance the robustness of models. Finally, a generative-aware conceptual framework is proposed at the end of the review to inform future research on the development of offline handwriting signature forgery detection system that is more resilient and forensic-ready.

A Web-Based System for Managing Student Attendance and Assessment Submissions: An SDLC Waterfall Model Approach

Mon, 01 Sep 2025 00:00:00 +0000

University lecturers handle numerous tasks, including tracking student attendance, managing multiple assessment submissions, handling online tests and quizzes, as well as providing timely evaluations of assessments. Currently, these processes are manually managed, which can be time-consuming and inefficient due to the absence of an integrated web-based application system. For example, students often submit their assignments or project papers through cloud storage links provided by lecturers. However, the lack of standardization in file naming frequently forces lecturers to spend additional time renaming hundreds of files before they can begin the grading process. This paper presents a proposed solution in the form of an integrated web-based application system designed to automate and streamline critical tasks like attendance tracking, assignment submission, and evaluation, serving as a comprehensive, one-stop solution for lecturers. SDLC (System Development Life Cycle) Waterfall Development Approach was applied as a method of system construction. HTML and PHP were employed for its implementation and handling of diverse functionalities. To assess the system’s effectiveness, it was tested over one academic semester with the participation of 111 degree and diploma students at Universiti Teknologi MARA, Perlis branch. Findings revealed that the system significantly simplified the attendance logging process and streamlined assignment submissions, as perceived by the students. In parallel, lecturers reported that the system facilitated a more efficient grading process, allowing them to return assessments promptly and allowing students to access their marks quickly and easily. Overall, the integration of these automated modules helped reduce bureaucratic inefficiencies, allowing lecturers to perform their duties with greater efficiency and less stress. Positive feedback from students also highlighted a desire for this system to be expanded to other courses, alongside suggestions for further enhancements to improve functionality.

Bibliometric Analysis of Research on Firth Penalized Logistic Regression in Addressing Complete Separation

Nurul Husna Jamian, Ahmad Zia Ul-Saufie, Mohammad Nasir Abdullah (Author) — Mon, 01 Sep 2025 00:00:00 +0000

Complete separation in logistic regression leads to infinite estimates which prevents reliable inference. Firth's penalized likelihood method has emerged as a widely accepted and reliable solution that provides finite and more stable estimates. Despite its growing relevance, a thorough understanding of the global research on this topic remains limited. This study conducts a bibliometric analysis of trends related to complete separation in logistic regression using Firth penalized regression. Bibliographic data were retrieved from the Scopus database and analysed using Microsoft Excel and VOSviewer software. After applying inclusion criteria, nine journal articles published between 2012 and 2024 were identified through a structured search conducted on February 22, 2025. The findings reveal a small but growing body of literature, reflecting the emerging status of research on complete separation in logistic regression using Firth penalized regression. The results show an upward trend in publications, particularly from 2019 onward with the United States and Malaysia identified as the most productive countries. Influential articles contributed to methodological development and applications in health and transportation research. Keyword co-occurrence analysis identified thematic clusters in human studies, statistical modelling, and estimation techniques. These findings provide an overview of publication trends, collaboration networks, and research gaps which could support future methodological and multidisciplinary integration of Firth penalized regression.

A Fuzzy Analytic Hierarchy Process (FAHP) Approaches for Investment Decision-Making in Malaysia

Mon, 01 Sep 2025 00:00:00 +0000

Investment decisions are essential for achieving financial growth and stability. Young Malaysians, however, often face challenges such as limited financial resources, rising living costs, and low financial literacy. Common investment options in Malaysia include gold, stocks, property, and cryptocurrency. Each option differs in capital requirements, profitability, risk level, and long-term viability. Selecting the most appropriate investment is difficult because decision making is often subjective and influenced by uncertainty. Conventional multi criteria decision making (MCDM) methods such as Simple Additive Weighting (SAW), Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS), and Analytic Hierarchy Process (AHP) have been widely used, but they are less effective in dealing with this challenge. To overcome this limitation, this study applies the Fuzzy Analytic Hierarchy Process (FAHP), which integrates fuzzy logic with AHP to capture expert evaluations more realistically. Four investment alternatives are assessed based on capital, profit, risk, and sustainability. The results indicate that gold is the most preferred investment option, followed by property, stocks, and cryptocurrency. By applying FAHP to investment decision-making in Malaysia, this study introduces a novel framework that not only supports young investors in making strategic choices but also offers insights that may inform financial literacy programs and policy initiatives under uncertain market conditions.

Assessing Knowledge Acquisition and Perceptions in Hands-On Video Editing Workshop for Non-Technical Students

Mon, 01 Sep 2025 00:00:00 +0000

In the era of digital technology, video editing is now a priority skill in various fields like education, marketing, and business. Nevertheless, non-technical students tend to lack video editing skills because of their limited exposure to montage elements and video editing software. This research explores the effects of practical video editing workshops on knowledge and perceptions of non-technical students. The workshops were aimed at imparting knowledge of montage elements, a key component of video production. With a quantitative research design, 30 students from non-technical disciplines have participated in guided workshops with Microsoft PowerPoint and CapCut. The findings indicate a significant improvement in the comprehension of montage elements, including visual organization, audio synchronization, transitions, and pacing among the students, with high mean knowledge scores (3.97 - 4.05) on a 5-point scale. In addition, regression analysis shows a significant relationship (p-value < 0.05), confirming its effectiveness in engaging non-technical students and achieving a better understanding of video editing concepts. The results of this research contribute to the development of instructional strategies that promote improved learning experiences for groups of students with varying backgrounds.