UTAR Institutional Repository (Universiti Tunku Abdul Rahman)
Not a member yet
6132 research outputs found
Sort by
System with display and advisory on payment technology preference through modern networking
Get PDFIn the modern evolving digital economy, businesses are facing increasing pressure to accommodate multiple wireless payment methods—such as digital wallets, credit cards, and cryptocurrencies—within their current enterprise infrastructure. However, the lack of integration among these systems typically results in business inefficiencies, security vulnerabilities, and siloed customer experiences. The purpose of this project is to address these problems by establishing a single, unified platform that integrates the wireless payment solutions into the Odoo Enterprise Resource Planning (ERP) system. The solution uses Stripe for processing digital wallet transactions and CoinGate for cryptocurrency payment processes, respectively, with real-time synchronization with critical business modules like sales, inventory, and accounting. A user-friendly dashboard has been developed that displays transactions and provides strategic payment recommendations using business intelligence (BI) approach. The system design, which is driven by Odoo's modular framework, offers secure payments, automation, and expandability to allow for future enhancements. By providing backend automation along with advisory visualization, the system provides enhanced financial intelligence, reduces errors, and encourages informed decision-making. Not only does this project ease payment management, but it also facilitates the greater good of establishing secure, smart, and efficient digital payment ecosystems for next-generation businesses
Predictive maintenance for server failure in virtual environments
Get PDFThis project develops a predictive maintenance framework for server failures in virtual
environments through a six-stage workflow. Large-scale datasets from Google Cluster,
Backblaze HDD, CINECA M100, and Azure VM traces were modeled to establish domainspecific
baselines, followed by the design of a unified schema enabling cross-domain
integration and transfer learning. Evaluation confirmed CPU stress and thermal load as
dominant predictors, with thresholding guided by operational risk. A simulation tool with
calibrated mathematical risk models, probabilistic scoring, and interactive dashboards was
implemented to address deterministic prediction issues, and a pseudo real-time system was
demonstrated using streamed logs. The workflow delivers a complete, simulation-ready
predictive maintenance pipeline with both academic rigor and practical deployment value
Blockchain-based product authenticity verification
Get PDFThis project proposes a blockchain-based product authentication system designed to enhance security, traceability, and transparency in supply chains, addressing the growing problem of counterfeit products in sectors such as pharmaceuticals, food, and electronics. Traditional centralised systems are prone to manipulation and lack transparency, resulting in counterfeit goods entering the market, which poses significant risks to consumers and businesses alike. The proposed system integrates blockchain technology with smart contracts to provide a decentralised, tamper-proof method for verifying product authenticity. Blockchain ensures that product data, including production and distribution history, is recorded immutably, while ECC offers efficient encryption for securing sensitive data. Additionally, the use of Ganache is utilised for decentralised data storage, ensuring data redundancy and security. The system incorporates blockchain details to enable secure consumer verification of product authenticity in real time. The methodology involves system architecture design, blockchain node setup, smart contract deployment, and local website development for verification purposes. The expected outcome of this project is an innovative, robust system that improves supply chain integrity, prevents the circulation of counterfeit products, and enhances consumer trust in product authenticity
Fall detection using gait analysis
Get PDFThis project focuses on fall detection for elderly populations using deep learning and gait
analysis. Falls are a major concern in aging populations, often resulting in severe injuries and
diminished quality of life. Traditional fall detection systems have limitations in accurately
identifying falls and adapting to real-world environments.
This study implements and compares five deep learning architectures: CNN, BiLSTM, BiGRU,
CNN-BiLSTM and CNN-BiGRU with attention mechanisms to capture both spatial and
temporal patterns in human gait. MediaPipe extracts pose landmarks from video frames, while
OpenCV aids in frame processing. The research process involves data collection from the
Multiple Cameras Fall Dataset with comprehensive preprocessing including two-stage
normalization and time-series scaling. Eight gait features are extracted: stride length, knee
angles, body velocity, acceleration, step frequency, posture angle, and arm swing. The data is
split using an 80/10/10 sequence-level approach to ensure models are tested on completely
unseen video sequences.
Performance evaluation using accuracy, precision, recall, F1-score and AUC-ROC metrics
shows the hybrid CNN-BiLSTM model achieving 90.04% test accuracy with 94.51% precision
and 94.92% recall on completely unseen data. Permutation-based feature importance analysis
reveals that arm swing is the most critical predictor across all models, followed by stride length
and knee angles. The MediaPipe approach demonstrates a 3.8-fold improvement in processing
time compared to traditional raw frame processing while maintaining detection accuracy.
The novelty lies in the systematic comparison of five architectures using sequence-level data
splitting to prevent data leakage, comprehensive feature importance analysis across multiple
models, and the fusion of real-time gait analysis with deep learning techniques. Results
demonstrate the hybrid model's ability to detect falls with high accuracy and minimal false
detections, providing an efficient and adaptable solution for fall detection in elderly care
settings
Comparing machine learning techniques to segmentize and classify tongue regions for traditional and complementary medicine (TCM) diagnosis
Get PDFThis project investigates the application of machine learning and deep learning techniques for automated tongue diagnosis in the context of Traditional Chinese Medicine (TCM). Tongue diagnosis, a long-established diagnostic method in TCM, is often limited by subjectivity and inconsistency. To address this, the study develops a systematic pipeline that integrates segmentation and classification models, enabling more objective, accurate, and reproducible analysis of tongue images. Three datasets—binary (stained vs. non-stained moss), four-class (color variations), and five-class (coating categories)—were utilized to evaluate performance under varying levels of complexity. Segmentation was performed using both classical methods (SVM) and a deep learning approach (DuckNet), with DuckNet providing superior accuracy and robustness. Classification was carried out through an evolutionary series of architectures, beginning with AdderNet and progressing through ResNet20, HybridNet, and an Improved HybridNet. Experimental results demonstrated that while AdderNet achieved the highest accuracy in complex multi-class scenarios, it suffered from excessive computational cost and scalability limitations. The Improved HybridNet consistently offered the best trade-off between performance and efficiency, delivering strong accuracy with reduced parameters, training time, and model size. Overall, the project highlights the potential of artificial intelligence to modernize tongue diagnosis by providing standardized, efficient, and clinically relevant computational tools. The findings establish a foundation for future integration of AI-driven diagnostic support systems into healthcare practice
Harnessing emotions using language processing in detecting cyberbullying
Get PDFCyberbullying represents a widespread challenge across social media platforms, frequently resulting in considerable emotional and psychological distress for individuals. Although current detection systems are geared towards recognizing harmful language, they fall short in comprehensively understanding the emotional consequences of such content. This project introduces Advanced Emotion Detection, a system aimed at categorizing particular emotions and assessing their intensity within comments on social media. This project centers on Instagram, a platform characterized by visual and textual interactions that often result in cyberbullying, with the objective of refining a large language model (LLM) by utilizing data obtained from publicly available posts and comments. The final dataset collected will be processed and used to fine-tune an LLM to locate subtle expressions of emotions within text. The system will go further than the basic sentiment analysis in detecting the severity and type of emotional impact, thus allowing the correct cyberbullying incident classification. The outcome of the project is to create a fine-tuned LLM model that capable to detect and classify the severity and types of cyberbully emotional impact
Trust, commitment, and conflict resolution styles as predictors of romantic relationship satisfaction among emerging adults in Malaysia
Get PDFMaintaining good romantic relationships is a critical developmental challenge during emerging adulthood, yet many emerging adults find it difficult to establish long-lasting and happy relationships. This study aims to explore how trust, commitment, and constructive and destructive conflict resolution styles influence relationship satisfaction among emerging adults in Malaysia. A quantitative, cross-sectional design was employed with 98 participants aged 18–25 years, recruited using purposive sampling. The sample comprised 56.1% females (n = 55) and 43.9% males (n = 43), encompassing Malay, Chinese, and Indian ethnicities. Standardized instruments, including the Relationship Assessment Scale (RAS), Dyadic Trust Scale (DTS), Dedication Subscale of Commitment Inventory (CI), and the Conflict Resolution Styles Inventory (CRSI), were administered. Data collection was conducted through online platforms using Qualtrics, and data were analyzed with IBM SPSS version 24. Multiple linear regression analysis revealed that trust (β = .452, p < .001) and commitment (β = .174, p = .049) significantly predicted higher relationship satisfaction, whereas destructive conflict resolution style (β = −.242, p = .004) significantly predicted lower relationship satisfaction. In contrast, constructive conflict resolution style was not a significant predictor (β = .064, p = .446). These findings align with Interdependence Theory, which emphasizes the importance of dependence and correspondence of outcomes in shaping relationship satisfaction. By highlighting the developmental and cultural factors influencing young adults’ relationships in Malaysia, the study contributes to the literature and provides practical implications for relationship education, counseling, and youth development initiatives
Healthy lifestyle management system
Get PDFThis project presents the design and development of a web-based healthy lifestyle management system with the combination of food, hydration and sleep tracking into a single platform. The motivation for this system arises when many individuals are aware of the importance of healthy diet, adequate plain water intake and sufficient of sleep hours but due to environmental influences and personal priorities, these practices are often neglected. This project is therefore proposed with the primary objectives that is to enable users to record personal information, track daily progress and receive daily recommendations tailored to their demographic profiles and lifestyle goals. This system was built using React with Typescript for the frontend to ensure a responsive and interactive interface while JSON Server was adopted as a lightweight backend solution to simulate database operations. The system features three main components which are food monitoring, hydration management and sleep tracking. Each module allows customization to suit different user objectives such as improving dietary balance, maintaining hydration or better sleep patterns. System testing was carried out through self-evaluation, focusing on functional correctness, data handling and accuracy of recommendation logic. The testing outcomes confirmed that the system performed as intended, with all core modules meeting the specific requirements which demonstrating reliability in storing, processing and displaying user data. In conclusion, this study highlights the potential of web-based applications in promoting healthier lifestyle practices. While current implementation serves as a prototype, it establishes a strong foundation for future enhancements. Potential enhancements may include integrating clinically verified food databases, applying the Beverage Hydration Index (BHI) to record hydration levels more accurately across different beverage types and age groups, and implementing Voluntary Application Server Identification (VAPID) push notification for daily reminder through web browser followed by implementing password recovery feature in the login page
Trust, subjective norm and attitude towards the purchase intention of green products among Generation Z
Get PDFThe objective of this research is set to understand trust, subjective norm and attitude towards the purchase intention of green products among Generation Z. In this research, the independent variables are trust, subjective norm, and attitude, while determining their direct relationship with dependent variable, purchase intention. The scope of our targeted sample size and subject to study is among the generation z in Malaysia, and 171 questionnaire respondents are successfully collected. With the aid from Statistical Package for Social Science (SPSS) Software to run the reliability tests, explaining the correlation coefficient, and testing of each hypothesized relationships between the existing variables. After experimenting and analysis, the results shown from Pearson Correlation Coefficient and Multiple Linear Regression Analysis determine the significant positive relationship between two of the existing independent variables (subjective norm, attitude) and our dependent variable (purchase intention) and insignificant positive relationship between one of the existing independent variables (trust) and our dependent variable (purchase intention). Keywords: purchase intention; trust; subjective norm; attitude; green products; Generation Z Subject Area: HF5413.W55 20
Factors influencing employee turnover intentions in Malaysia's aviation industry
Get PDFEmployee turnover is an ongoing challenge in Malaysia’s aviation industry, where retaining skilled and experienced employees is crucial for efficiency, service quality, and safety. This study aims to examine the factors influencing employee turnover intention by focusing on job embeddedness, organizational citizenship behaviour (OCB), and emotional exhaustion. A quantitative research method was employed through structured online questionnaires distributed to employees in Malaysia’s aviation sector, including Malaysia Aviation Group and Capital A. Findings revealed that job embeddedness and OCB are significantly and negatively related to turnover intention, indicating that employees with strong workplace ties and discretionary actions are less likely to leave. In contrast, emotional exhaustion shows a significant positive relationship with turnover intention, highlighting that prolonged stress and resource depletion increase the likelihood of resignation. This research contributes theoretically by exploring how job embeddedness, OCB, and emotional exhaustion influence turnover intention in Malaysia’s aviation industry. Practically, it provides valuable insights for aviation companies in designing effective retention strategies. These include fostering supportive work environments, encouraging citizenship behaviours, and implementing well-being programs to reduce burnout. Addressing these factors can help strengthen workforce stability, enhance efficiency and service quality, and support the long-term sustainability of aviation companies in a competitive industry. Keywords: Employee Turnover, Job Embeddedness, Organizational Citizenship Behaviour (OCB), Emotional Exhaustion, Aviation Industry Subject Area: HD4801-8943 Labou