UTAR Institutional Repository (Universiti Tunku Abdul Rahman)
Not a member yet
6132 research outputs found
Sort by
论韩愈干谒书信的自我书写 : The self-expression of Han Yu’s self-recommendation letters
Get PDF干谒书信不仅反映了干谒者对进入仕途的渴望，也体现出当时社会的荐举 制度与士人求仕的困境。本文使用了文本分析、文献研究、比较研究与历史批 评等研究方法。通过分析韩愈 11 篇干谒书信，结合前人研究成果与历史文献， 再将其与韩愈其他作品进行对比，并结合当时的时代背景来深入剖析韩愈干谒 书信中“复古求奇”与“人微贫苦”的自我形象建构。接着，结合韩愈的其他 作品与事迹来探讨他在不同阶段对干谒行为的心态演变。韩愈在书信中塑造出 复古、求奇的形象，也描写自身生活之困顿，希望获得干谒对象的关注与同情， 进而被举荐和提拔。韩愈对干谒行为的态度也历经转变，从最初的排斥、羞愧 到逐渐接受，并最终肯定其合理性与必要性，甚至鼓励后进之士行干谒以施展 才能。学界关于韩愈干谒书信的研究主要集中在探讨其心态转变，关于研究自 我书写的研究较少。本文探讨了韩愈生平、求仕思想之形成与其干谒书信创作 背景，再分析信中的自我书写，后再进一步对比分析信中所体现的对干谒之看 法，系统性地对韩愈干谒书信的自我书写与干谒心态进行研究。
【关键词】韩愈、干谒、求仕、干谒书信、自我书
Exploring the impact of social media marketing on consumer brand engagement in fashion branded jewellery
Get PDFSocial media marketing has changed the way brands connect with customers, especially in the fashion branded jewellery industry in Malaysia. This study will use the timulus Response Theory to understand how brand’s marketing efforts on social media influence consumer engagement. The study focuses on social media marketing activities like entertainment, interaction, electronic word-of-mouth and trendiness as the "stimuli" and looks at how these lead to consumer brand engagement as the "responses." Data will be collected from 195 respondents through surveys and analyzed
using SPSS software. This study can provide useful ideas for marketers to improve their social media strategies and stay ahead in the competitive fashion branded jewellery market. The research also will offer guidance for future studies to better understand how social media marketing affects consumer brand engagement.
Keywords: Social Media Marketing, Stimulus-Response Theory, Consumer Brand Engagement, Fashion Branded Jewellery, Consumer Behavio
Investigating Malaysia youth's preferences: long form vs short form product review video when considering high involvement product
Get PDFWhen purchasing a high-involvement product, which is usually identified by its greater price and great personal
significance, careful research and trust are necessary. The purpose of this research is to find out which type of
product review videos youths in Malaysia prefer, long or short-form, and what factors influence those choices.
The study utilises a quantitative research design and distributes an online survey to fifty participants aged between
15-40 years old, who spend a lot of time in a digital context, intend to purchase a high-involvement product, and
have gotten in touch on product review videos via social media platforms using Google Forms. To gain insight
into the watching preferences of the intended audience, the Statistical Package for the Social Sciences (SPSS) is
used to analyse the obtained data. Initial findings reveal that perceived depth of understanding, clarity of product
feature demonstration, helpfulness in resolving doubts, credibility, effectiveness of product comparison,
presentation of pros and cons, addressing product concerns, emotional attachment, level of assurance, influence
on post-purchase regret, ability to reduce post-purchase regret, reduction in post-purchase financial concerns, and
increasing post-purchase confidence significantly influence Malaysian youths’ preferences toward long-form
versus short-form product review videos when considering high-involvement products. Marketing professionals
and influencers can create compelling product review content that appeals to Malaysia's youth consumer base by
taking into account these preferences and considerations. The limitations in this study are that the Wilcoxon
Signed-Rank Test only compares the rank of differences instead of the actual size of the differences, the presence
of ties may slightly reduce the strength of the statistical results, and the sample size is limited to Malaysian youths.
Hence, the finding may not be generalised to older consumers. For future recommendations, the research can
include a more diverse sample with different age groups, income levels, and geographic areas. Investigation of
consumers' preferences on video formats across different mediums can also be considered with different research
instruments, such as interviews or experiments, to get a deeper insight into why consumers trust certain video
formats more than others.
Keywords: consumer decision-making process; high-involvement product; low-involvement product; product
review video; long-form video; short-form video,
Subject Area: HF5410-5417.5: Marketing. Distribution of product
Sentiment analysis of financial news for predicting stock price trends using NLP techniques in fintech
Get PDFThe stock market is highly influenced by news and investor sentiment, making trend prediction
both challenging and valuable. This project develops a framework for stock price trend
prediction by integrating sentiment analysis of financial news with historical market data. News
headlines are cleaned and analyzed using VADER, TextBlob, BERT, and FinBERT to generate
sentiment scores, which are merged with OHLCV price data and enriched with lagged returns
and time-based features. Five machine learning models — Support Vector Machine (SVM),
Logistic Regression (LR), Random Forest (RF), XGBoost, and LightGBM — are trained and
tuned using walk-forward cross-validation. Their performance is evaluated using accuracy,
precision, recall, F1-score, and confusion matrices, with XGBoost achieving the best results.
Finally, a Power BI dashboard is built to visualize sentiment trends, market data, and model
predictions, making insights interactive and actionable. Results show that incorporating
sentiment features improves predictive performance, supporting data-driven decision-making
for investors and analysts
Smart home monitoring in Node-RED emulator using Flutter mobile app and AI technology
Get PDFThe limitations of static automation in current IoT-based smart home systems highlight
a lack of flexibility and personalization. To address this gap, this project develops an
adaptive smart home control and monitoring system that integrates Internet of Things
(IoT) and Artificial Intelligence (AI) technologies for dynamic automation. The system
is emulated using Node-RED with an SVG-based interface for device visualization,
while a Flutter mobile app serves as the user interaction platform. MQTT provides realtime
communication, and InfluxDB supports time-series data management. An AI
module leverages both historical and real-time data to enable predictive decisionmaking,
enhancing functions such as temperature regulation, lighting management,
and air purification. The prototype demonstrates that adaptive automation improves
comfort, responsiveness, and energy efficiency compared to traditional static methods.
Project deliverables include the partial development of the Flutter mobile UI, Node-
RED flows, and the SVG emulator, with future work focusing on full integration of
InfluxDB for advanced analytics and AI-driven predictive control through machine
learning
Raspberry Pi–based cyber-physical system for e-bike monitoring over the internet
Get PDFThis project addresses the growing problem of e-bike theft by developing a Raspberry Pi-based
Cyber Physical System (CPS) for real-time monitoring and security. With e-bike theft rates significantly exceeding motor vehicle theft and recovery rates below 15%, innovative anti-theft technologies combining prevention and recovery capabilities are urgently needed. The system integrates facial recognition using the InsightFace framework, motion detection via MPU6050
sensors, and GPS tracking with the NEO-6M module. The Raspberry Pi 4 Model B coordinates sensor data through an event-driven architecture, enabling real-time threat assessment and
immediate response activation. A Logitech C270 webcam provides biometric authentication,
distinguishing between authorised and unauthorised users with a confidence-based assessment.
MQTT communication protocols via EMQX broker ensure reliable data transmission to a cross-platform Flutter mobile application. The mobile interface provides comprehensive monitoring through facial recognition results, interactive GPS tracking with route visualisation,
and system status monitoring. Security events automatically trigger GPS tracking activation
and mobile notifications for immediate theft response. Testing in Kampar, Malaysia,
demonstrated reliable performance with successful unknown face detection, automatic security
response activation, and accurate location tracking. The system achieved effective integration
of biometric authentication, motion sensing, and location tracking into a unified security
platform, providing automated theft detection and asset recovery capabilities essential for ebike protection in urban environments
AI-driven multi-class classification of X-ray images for disease detection
Get PDFThere are critical challenges in medical imaging diagnostics, including human error in radiological analysis, slow diagnostic processes, and radiologist shortages. This project developed an AI-driven multi-class classification system for chest X-ray disease detection. The research employed a DenseNet121 architecture with transfer learning to classify 13,482 chest X-ray images into five categories: COVID-19, pneumonia, tuberculosis, lung opacity, and normal cases. The methodology implemented a two-stage progressive unfreezing strategy, comprehensive data preprocessing with augmentation techniques, and class weight balancing to address dataset imbalances. Training utilized TensorFlow and Keras frameworks with GPU acceleration, incorporating early stopping and learning rate reduction callbacks for optimization. Gradient-weighted Class Activation Mapping (Grad-CAM) was integrated for AI interpretability, and a comprehensive Streamlit dashboard was developed featuring real-time processing capabilities. The DenseNet121 model achieved exceptional performance with 94.52% validation accuracy, 94.7% test accuracy, 95% precision and recall, and 0.99 AUC score across all disease categories. The system successfully demonstrated clinical-grade interpretability through visual attention mapping and deployed as a functional web application with hospital finder and weather health monitoring features. This research establishes a foundation for AI-assisted medical diagnosis, potentially improving healthcare accessibility and diagnostic reliability while maintaining transparency in AI decision-making processes for clinical deployment
Development of an interactive web-based platform for SQL learning and skill enhancement
Get PDFThis project outlines the creation of SQL Quest, an interactive platform that aims to assist in learning SQL through gamification. SQL, or Structured Query Language, is essential for data management and analysis, however, most online learning platforms are not engaging, lack personalized feedback, or do not provide real-world relevance. SQL Quest seeks to provide a solution by integrating conventional content into interactive practices coupled with real-time feedback, an XP progress tracker, and challenges to motivate learners. The system is built with HTML, CSS, and JavaScript alongside web development software such as Node.js and MySQL. These components structure the platform using the Prototyping model which allows for iterative refinement. The platform boasts a few features including lesson modules with quizzes, practice SQL editors, multi-tiered challenges, dashboards, and a progress tracker. Users are also enabled to bookmark lessons, track lessons they have completed, and earn XP which unlocks badges. SQL comprehension and user engagement preliminary results indicate that user functionality and performance meet SQL Quest’s objectives. The project has value in responsiveness, scalability, and user-centered design principles applied to SQL and computer science education. Further development five aims to incorporate a SQL Smart assistant for suggestion and correction tasks to enhance interactivity and incorporate industry-specific SQL application scenarios
Lab inventory monitoring system using low code programming
Get PDFThe area of study for this project is web application development. This project aims to develop an inventory monitoring web portal for solving the problems lacking on existing inventory monitoring systems.
Nowadays, the existing systems still have complex interfaces and navigation structures, which causes time consuming and high cost in adapting this kind of system. Moreover, there are many business enterprises still rely on paper works and excel tables in monitoring inventory. This causes a lot of external error occurring during inventory monitoring and affects the accuracy of monitoring and reporting which resulting below standard level. Additionally, the existing systems don’t provide further general instruction for addressing abnormal items. This will cause the unnecessary workflows to be repeated persistently and reduce the efficiency of operation.
Furthermore, the proposed inventory monitoring web portal is developed by using low-code programming, which is Node-RED. Then, this web portal is integrated with database and various APIs which consists of Telegram chatbot API, MySQL database, and so on. Besides that, the process of development of inventory monitoring web portal follows the standards of Agile model.
Then, this project has also conducted literature review on 4 existing inventory monitoring systems for finding out their limitations. The proposed system allows improve the accuracy and efficiency in monitoring inventories.
Consequently, this project develops an inventory monitoring web portal by utilizing the low-code development tool to address the key challenges faced by the existing systems
Analyzing the impact of corporate social responsibility (CSR) on purchase intention of Generation Z in fast-food industry
Get PDFThis study examines the effect of Corporate Social Responsibility (CSR) on purchase intentions among Malaysian Gen Z in the fast-food industry, using Carroll’s four CSR dimensions (economic, legal, ethical, philanthropic). A qualitative analysis of 400 respondents (18–28 years) from Selangor, Johor, and Perak via quota sampling was conducted. Multiple regression results show ethical responsibility (β = 0.506, p<0.001) is the strongest predictor, followed by economic responsibility (β = 0.284, p=0.003). Legal (p=0.122) and philanthropic (p=0.518) were insignificant, suggesting Gen Z prioritizes ethical practices and fair pricing over compliance or charity. The model explained 37% of variance (R² = 0.37), indicating other influencing factors exist. Findings highlight Gen Z’s preference for CSR integrated into business operations rather than philanthropy. Fast-food brands can boost engagement through ethical sourcing and fair pricing. Limitations include urban sampling bias and self-reported data; future studies should cover broader demographics and experimental methods for stronger causation. Keywords: Corporate Social Responsibility, Purchase Intention, Generation Z, Fast-Food Industry, Ethical Responsibility, Economic Responsibility, Legal Responsibility, Philanthropic Responsibility Subject Area: HD60-60.5 Social responsibility of busines