UTAR Institutional Repository (Universiti Tunku Abdul Rahman)
Not a member yet
6132 research outputs found
Sort by
Multi-fuzzer techniques for automated vulnerabilities assessments
Get PDFIn the contemporary cybersecurity landscape, effectively safeguarding software from vulnerabilities is critically important. This project introduces an innovative approach to automated vulnerability assessment through a sophisticated multi-fuzzer system designed to enhance the security of at-risk software applications. The primary objective is to provide an efficient and user-friendly solution for identifying and analyzing security vulnerabilities via a dynamic front-end chatbot interface. Users can seamlessly upload their software applications, which are subsequently subjected to a series of diverse fuzzing tools within an automated framework. The system employs a range of fuzzing tools, such as AFL++ and Honggfuzz, ensuring a comprehensive and systematic evaluation of software interfaces and their responses to various potential threats. By automating the fuzzing process, this project facilitates a more efficient and thorough assessment of security weaknesses than traditional manual testing methods. The automated framework generates detailed CVEs on discovered vulnerabilities and potential exploitation scenarios, significantly enhancing the security posture of the evaluated applications. The results of this project demonstrate the system's capability to automatically detect and document vulnerabilities across different software environments, providing a comparative analysis of the effectiveness and limitations of various fuzzing techniques. This analysis offers valuable insights into the roles these techniques play in software security, highlighting the importance of using a multi-fuzzer approach to achieve a more resilient vulnerability assessment. Ultimately, this project underscores the critical role of automation in vulnerability assessment and reinforces the value of employing diverse fuzzing methods as essential tools in advancing cybersecurity practices. The findings contribute to the development of more effective security measures and serve as a foundational resource for improving software security in future applications
Home surveillance in general
Get PDFThis project addresses a growing concern in modern society – home security. With increasing incidents of property crime and unauthorized intrusions, there is a rising demand for intelligent surveillance systems that go beyond the limitations of conventional CCTV setups, which often struggle with false alarms and require manual supervision. This project proposes a smart home surveillance system that combines real-time object detection with violence recognition by leveraging state-of-the-art deep learning techniques.
The system uses the YOLO (You Only Look Once) framework to detect the presence of weapons, offering rapid identification of potential threats. Simultaneously, a ResNet50-based Convolutional Neural Network (CNN) combined with a Long Short-Term Memory (LSTM) network is employed to recognize violent actions over time, such as assaults or robberies, using temporal video frame analysis. When a human is detected in the scene, these detection modules are triggered to identify weapons or violent movements. If a threat is confirmed, the system issues an immediate alert to property owners or security personnel, enabling quick intervention.
By integrating real-time weapon and violence detection in a multi-threaded monitoring system, this solution enhances home surveillance effectiveness and responsiveness, aiming to create a safer and smarter living environment
Automated exploration data analysis tool
Get PDFThis project focuses on developing a web-based automated EDA tool This tool designed to simplify and streamline the EDA process. The primary objective is to automate the labour-intensive manual EDA procedures, eliminating the need for deep statistical or programming knowledge. The project addresses the limitations of existing tools, particularly for analyzing textual data by enhancing text data exploration capabilities. The system involves building a user-friendly interface using HTML for the front-end and Flask for back-end processing. Users can upload datasets and receive immediate visual and statistical analysis, including correlation heatmaps, bar charts, word clouds, and other visualizations tailored to numerical, categorical, and text data types. Text preprocessing features, such as tokenization and stop word removal are incorporated to handle textual data more effectively. The system will provide automatic insights based on dataset characteristics to reduce human error in data exploration. In summary, the tool can democratise data analysis by lowering the time and effort required for data preprocessing and visualisation while also making it available to a wider audience
An analysis on native chicken meat production in Malaysia: the development of chicken growth performance, feed efficiency and profitability framework based on different diet systems
Get PDFIn Malaysia, the poultry industry is a fundamental pillar of livestock production, contributing significantly to the nation’s agricultural economy. This study focuses on the native chicken production, a sector gaining increasing attention due to its lower fat content, enriched Omega-3
composition and highly regarded flavour profile. However, native chicken rearing is characterized by longer production cycles compared to conventional broilers. Additionally, the poultry industry in Malaysia faced significant challenges following the pandemic, with escalating chicken feed costs
impacting production profitability. To address these issues, optimizing feed strategies becomes essential to reduce production costs and mitigate risks associated with currency fluctuations for imported feeds, ultimately bolstering
the prospects of local poultry farmers.
The general objective of this study is to develop a framework of chicken growth performance, feed efficiency and profitability for native chicken meat production based on different diet feed systems in Malaysia with four Specific
Objectives: (1) to describe the different diet feed systems for native chicken meat production; (2) to determine the feed efficiency of the six (6) different diet feed systems based on the cumulative voluntary feed intake (CVFI);
cumulative weight gain (CWG); average daily gain (AVG) and feed conversion efficiency ratio (FCE) for the native chicken meat production; (3) to analyse the relationship among the six (6) different diet feed systems, chicken age and body weight for native chicken meat production; and (4) to examine the profit level using the benefit-cost ratio (BCR) of the six (6) different diet feed systems for the native chicken meat production. To carry out this research, the study was conducted at Bintang Maju Agri Chicken Farm
in Semenyih, Malaysia, involving the utilization of six (6) different diet feed systems. These encompass (1) Premium Starter Feed in 50kg pack (as control feed) (2) 5% Pokok Ketum Ayam (Trichanthera Gigantean) mixed into Premium Starter Feed (3) 5% Protein Larva Askar Hitam (BSFL) mixed into Premium Starter Feed (4) 200g Crude Palm Kernel Oil (CPKO) mixed into 50kg Premium Starter Feed (5) 100g Organic Acid mixed into 50kg Premium Starter Feed and (6) 100g Yellow Pigment (Brand: SK Gold) mixed into 50kg Premium Starter Feed. A total of 300 chickens, distributed across these six diet systems, underwent two production cycles, each spanning 84 days or 12 weeks.
The study comprised four fundamental phases. Initially, the selection of diet feed systems was grounded in a comprehensive literature review and consultation with poultry farmer and feed providers. Subsequently, feed
efficiency was meticulously assessed using various parameters, including cumulative voluntary feed intake (CVFI), cumulative weight gain (CWG), average daily gain (AVG) and feed conversion efficiency ratio (FCE). Following this, quantitative analyses, employing regression and panel data techniques via Eviews 12.0 software, were carried out to elucidate the relationships between the diet feed systems, chicken age and body weight. Finally, profitability was gauged through the benefit-cost ratio (BCR), involving a comparison of revenue and production costs.
The study’s findings revealed that diet feed system 6 (Yellow Pigment), consistently emerged as the top performer, yielding the highest growth performance across the study’s parameters. Simultaneously, diet feed system 5 (Organic Acid) and diet feed system 6 exhibited the highest feed efficiency, with FCE ratio of 2.469 to 2.533. The panel data analyses revealed noteworthy insights: diet feed systems 5 and 6 exhibited significant positive relationships with chicken body weight, suggesting a favourable impact on growth (p<0.05). Conversely, diet feed systems 3 (Protein Larva Askar Hitam) displayed a significant negative relationship with chicken body weight, compared among
all diet feed systems (p<0.05). For profitability, all diet feed systems demonstrated better profitability compared to diet feed system 1 (control feed), that is with benefit-cost ratio exceeding that of the control feed. Diet feed system 6 stood out as the most cost-effective choice over two production cycles spanning 12 weeks with benefit-cost ratio of 1.38 to 1.40. The study highlights the importance of strategic feed formulation in boosting poultry growth, improving feed efficiency and maximizing profitability — critical elements for scaling up sustainable and competitive native chicken meat production. The findings of this study offer clear, actionable insights for poultry farmers, feed manufacturers and agri-entrepreneurs seeking to improve
production efficiency and profitability using diet feed system 6 (Yellow Pigment) and diet feed system 5 (Organic Acid).
Keywords: Native chicken meat production, chicken growth performance, feed efficiency, profitability, diet feed systems
Subject Area: SF94.5-99 Feeds and Feedin
Developing an IoT-based inventory management system using Node-RED: Enhancing smart factory environments
Get PDFThe integration of the Internet of Things (IoT) within smart factory environments is revolutionizing inventory management processes to make them more efficient and responsive. This project focuses on developing a simulation-based IoT inventory management system using Node-RED to demonstrate smart factory operations such as real-time monitoring, inventory tracking, predictive maintenance, and quality control. The system integrates technologies of Node-RED, MQTT, InfluxDB, and Telegram to simulate and visualize factory activities.
All data used in this simulation is generated virtually within Node-RED to mimic the behavior of a real-world manufacturing environment. This includes simulated temperature, humidity and pressure readings, electronic components movement across various stages (mounting, soldering, inspection, testing, packaging), inventory level changes, and machine runtimes. The system also features a notification mechanism via Telegram to alert factory managers in real time about key events such as high temperature, defective components, low inventory levels, and scheduled maintenance.
This simulation provides a cost-effective and scalable platform for testing and visualizing IoT-based inventory management solutions. The use of Node-RED allows flexible flow-based programming so that the system is highly adaptable for educational and prototyping purposes. Additionally, real-time data tracking, automated decision- making, and visualization through the Node-RED Dashboard enable a deeper understanding of how smart factories can operate more efficiently.
This project highlights the potential of IoT integration for achieving improved resource utilization, reduced downtime, and enhanced decision-making in inventory management processes. It serves as a foundation for further development toward more intelligent, connected, and automated factory systems
Food ordering application
Get PDFThe demand for food ordering services has caused several problems in improving customer satisfaction and operational efficiency. This research presents a food ordering system that solves issues like food suggestion for people with allergies and specific dietary needs, delays in order processing, and the lack of ingredients in manage side. The system offers a tailored food ordering experience, focusing on user safety by providing allergy-sensitive meal suggestions, accurate preparation time estimates, and manage ingredients efficiently. The system can also improve kitchen workflows, reduce delays and improve customer satisfaction. Also, the integration of inventory management lets restaurant owner make early decisions about ingredients. This research helps develop smart, efficient, and clear food ordering systems, improving user experience without focusing on delivery and payment processes
AI-powered breed identification and personalized care
Get PDFThis project investigates the integration of deep learning and modern artificial intelligence to enhance mobile pet care applications. The proposed system employs the MobileNetV2 architecture, a lightweight and efficient Convolutional Neural Network(CNNs) , to automatically identify pet breeds from photographs captured or uploaded by users. To provide foundational breed information, the application incorporates data from Dog API by Kinduff, Dog CEO and API Ninjas Dogs, which offer structured trait ratings such as shedding levels and energy characteristics along with general breed care tips. These sources supply only broad, breed-specific guidance and are implemented as fixed content within the application. To deliver care recommendations that go beyond generic advice, the system integrates the Google Gemini large language model, allowing users to request personalised guidance that considers their own pet’s breed, age and medical records. This dual approach addresses key limitations of existing pet care applications, which often rely on manual breed entry and offer only general advice, by combining automated breed recognition with customised, context-aware care recommendations. As a result, the system reduces human error, improves the reliability and relevance of pet health recommendations and simplifies overall pet management, thereby promoting better pet well-being and enhancing user satisfaction
Courtxpert hub: court reservations & equipment platform
Get PDFThe CourtXpert Hub project addresses the critical gaps in existing sports court booking systems within Malaysia, particularly in Kampar, Perak. Despite the availability of booking services across Malaysia, current systems such as Courtsite, AFA, 7Stone, and IOI Sports Centre lack integrated Smart dashboa options, multilingual accessibility, and an advanced administrative dashboard. These shortcomings often lead to inconvenience for casual players, reduced inclusivity among multilingual users, and inefficient decision-making by administrators. To overcome these limitations, the proposed platform was enhanced with three major features: an equipment rental prompt, a smart administrative dashboard with predictive analytics, and multilingual support in English, Bahasa Melayu, and Chinese. The system was developed on a PHP–MySQL architecture with Stripe API integration for secure payments (Visa, FPX, and GrabPay). The methodology followed an Agile, incremental development approach, incorporating iterative design, testing, and refinement throughout the process. Moreover, system evaluation demonstrated reliable booking workflows, real-time equipment inventory management, and accurate predictive insights into revenue and venue utilization. In addition, administrators were also equipped with customer retention analytics and re-engagement tools, such as WhatsApp messaging integration, to improve user loyalty. As a result, the findings confirm that CourtXpert Hub successfully achieves its objectives by providing a comprehensive, inclusive, and intelligent booking platform. It not only improves user accessibility and convenience but also empowers administrators through data-driven decision-making tools. The project contributes to the digitalization of sports facility management in Malaysia and lays the foundation for scalable, nationwide deployment
Determinants of intention to use Islamic digital banking among Generation Z in Malaysia
Get PDFThe rapid growth of financial technology has transformed the banking industry, with Islamic digital banking emerging as an important innovation that aligns with Shariah principles. However, the intention of Generation Z to adopt Islamic digital banking in Malaysia remains a critical area of study, as this generation represents the future of financial service consumers. This study investigates the determinants influencing Generation Z’s intention to use Islamic digital banking, with emphasis on five major independent variables: performance expectancy, effort expectancy, social influence, facilitating condition, and awareness. A quantitative research design was employed, and data were collected from at least 384 respondents using a convenience sampling method. Questionnaires were distributed to obtain primary data, which were then analysed using the Statistical Package for the Social Sciences (SPSS) version 30.0 through Multiple Linear Regression Model Analysis. The findings provide evidence that these variables significantly affect the behavioural intention to adopt Islamic digital banking among Generation Z in Malaysia. The study highlights the importance of enhancing innovation, improving internet coverage area, and raising awareness to increase adoption rates. These insights can guide financial institutions, policymakers, and educators in promoting Islamic digital banking, ultimately contributing to a more inclusive and sustainable financial system in Malaysia. Keywords: Behavioural intention to use Islamic digital banking, performance expectancy, effort expectancy, social influence, facilitating condition, awareness, gen
Smart management system for tuition centre operations
Get PDFThis project, titled “Smart Management System for Tuition Centre Operations”, aims to develop a comprehensive solution to meet and fulfil the unique needs of small-sized tuition centres. The system addresses key challenges such as ineffective manual administrative and daily operational tasks, a lack of affordable management software, and poor communication between parents and tutors. This project prioritises rapid prototyping, iterative feedback from users, and continual refinement by utilising the Rapid Application Development (RAD) methodology. This ensures that the final product effectively satisfies user needs.
This system features a robust set of key modules, which include user authentication, student management, course management, class management with biometric facial recognition for attendance tracking, fee management, communication, and reporting and analytics. The system is created with Visual Studio Code (VS Code), Cloud Firestore Database, Firebase Functions, Firebase Authentication, Firebase Storage, Flutter, Dart, Node.js, JavaScript, and Python. The goals of this project are to increase communication between parents and tutors, streamline administrative and daily operational tasks, and boost the overall efficiency of tuition centres.
This project not only offers a practical tool for tuition centre management but also contributes valuable insights into the fields of educational technology and software development. The deployment of this system is expected to significantly reduce manual workload, reduce human errors, and foster better engagement between administrators, tutors, students, and parents