Your search

Formative assessment has long been recognised as a valuable educational measurement tool; however, its application in music education in Asia, particularly in Macau, remains underexplored. This study investigated the implementation of formative assessment in Macau and the factors that influence its application. A convergent mixed-methods research approach was adopted, utilising questionnaires, interviews and observations. Initially, teacher questionnaires, interviews and classroom observations were used to comprehensively understand how teachers implement formative assessment in singing teaching within junior middle schools in Macau. This approach identified the strengths and weaknesses of the practices while drawing insights from existing literature. Additional teacher questionnaires and interviews were used to explore the factors shaping teachers’ approaches to formative assessment. To provide a holistic perspective, student questionnaires and interviews were used to examine students’ experiences and perceptions of formative assessment in singing instruction. The findings indicate that a) teachers use all five formative assessment strategies outlined by Thompson and Wiliam (2007) with varying frequencies; b) teachers prioritise skill goals and technical accuracy over expressive qualities; c) performance assessment is common in singing classes, while questioning is more typical in other settings; d) teachers prefer teacher-directed assessments over student-directed ones (self and peer assessment); e) strengths include effective teacher demonstrations, frequent descriptive feedback, self-recording, reflective questions and constructive peer feedback, while weaknesses comprise infrequent use of assessment tools, application of multiple evaluative feedback types and low specificity in some descriptive feedback and misuse of self and peer ratings; f) both personal (teacher attitude, self-efficacy and subjective norms) and contextual (school environment, student challenges and public performances) factors hinder formative assessment implementation; and g) students value strategies like teacher demonstration, descriptive feedback and self-recording but devalue others such as evaluative feedback, self-rating, peer rating and peer feedback. Expert interviews were conducted to address the identified weaknesses and formulate targeted recommendations for stakeholders, aiming to enhance the effectiveness of formative assessment in singing teaching in Macau junior middle schools.

Artificial intelligence (AI) is changing the way we operate as a society. Generative AI models are especially known for being used to generate synthetic artifacts, such as texts, music, and images. This doctoral thesis explores generative AI's ability to create accurate images from prompt text. Our work aims to prove how generative AI tools are creating images that are remarkably similar in appearance as those created by humans. In addition to the theoretical contributions, this thesis explores broader secondary open questions about generative AI: what implications arise for the perception of what is virtual and non-virtual in our contemporary visual landscape? How does the new nature of interaction with generative AI change human-machine communication? Generative AI tools saw a series of breakthroughs these last years, which led to models that generate texts and images that are increasingly more difficult to distinguish from human- made creative content. As of 2022, Open AI developed and released ChatGPT, a chatbot enabling human users to converse, ask questions, explain concepts, and create new text-based content. However, the capabilities of generative AI went far beyond text generation. For example, gen AI models, Midjourney and DALL-E 3, are specifically designed to create images based on text prompts. These images are artificially created, meaning every screen pixel was produced using AI. Throughout this research, we explore new concepts of creative content generation, perception of virtual and non-virtual, memory, and trustworthiness in our contemporary imagery. Using an interdisciplinary methodological framework, this thesis engages with the creation of synthetic imagery as an opportunity for an infinite source of creativity or a detrimental disruption of contemporary visual culture.

Facial expression recognition is a key topic in computer vision, playing a crucial role in non-verbal communication. With the rapid development of artificial intelligence, significant progress has been made in improving recognition accuracy and generalization abilities. Traditional methods often suffer from low precision and poor generalization, while deep learning models have substantially advanced the field. However, deploying deep and complex models on different platforms remains challenging due to their high computational demands and frameworks. Hence, developing an efficient, real-time, and lightweight facial expression recognition system is critical. This study focuses on creating an efficient, accurate, and lightweight real-time facial expression recognition system with an emphasis on cross-platform deployment. It integrates various deep learning and optimization techniques to demonstrate flexibility across platforms. In this context, this study evaluates the performance of 10 advanced CNN models (VGG16, VGG19, ResNet50, etc.) on the facial expression dataset FER2013. YOLOv8 combined with ResNet50 achieved 70.56% accuracy on FER2013, outperforming YOLOv8 alone by 2.1%. Multi-module fusion models (MobileFaceNet, IR50, HyViT, SE) achieved an accuracy of 92.58% and 74.8% on the RAF-DB and FER2013 datasets, respectively, showing superior performance in ablation experiments. Given the significant impact of data quality on model performance, this study performed data cleaning on the FER2013 dataset, resulting in a 3.25% accuracy improvement for the YOLOv8 + ResNet50 model, which reached 73.81%. The high-resolution RAF-DB dataset, with fewer errors, led to improved performance, achieving 92.56% accuracy with the fusion model. A multi-purpose facial expression recognition system, VISTA, was developed using Python and PyQt5. The system supports multiple data formats and provides real-time emotional feedback, thus enhancing its usability for both research and practical applications. Furthermore, the fusion model was quantized using the OpenVINO toolkit, reducing its parameters by 75% while maintaining an accuracy of 91.17%. Inference speed was improved, and XGrad-CAM was employed to enhance model interpretability, revealing that the YOLOv8 + ResNet50 combination more effectively captured facial features. Finally, the high-performance model was successfully deployed on Intel CPUs, NVIDIA GPUs, and embedded devices Raspberry Pi 4B, demonstrating the portability and flexibility of the VISTA system across various platforms. This research provides promising solutions for applications in human-computer interaction, affective computing, and real-time emotional analysis, with significant advancements made in improving system real-time performance, accuracy, and cross-platform deployment capabilities. It contributes to the development of facial expression recognition technology and lays the foundation for its widespread future applications in fields such as smart healthcare, business analytics, education, and mental health.

This study investigates the regulatory-operational dynamics of the premium market in Macau casinos using a constructivist grounded theory approach. Qualitative in-depth interviews were conducted with 25 participants from three stakeholder groups. The grounded theory analysis identifies seven core categories illustrating the impact of the new regulatory system on the premium direct and premium mass segments in casinos. These categories include regulatory changes and operational challenges, decline of gaming promoters, premiumization of the gaming market, new player retention strategies, cross-border player acquisition risks, presence of unauthorized agents in casinos, and non-gaming development amid international competition. These insights highlight four major regulatory impacts on the industry themes and operational trends, i.e., industry regularization, market premiumization, product diversification, and criminal fragmentation. This study also identifies specific regulatory mechanisms and operational management in premium gaming, such as premium player identification, enhanced operational procedures, multi-tiered market segmentation, and the provision of personalized services. Additionally, stakeholder perceptions of the new gaming regulatory system are explored, with casino and expert groups considering it necessary and junket participants finding it restrictive. From these findings, a conceptual model structured around regulatory functions, compliance, and improvement is developed to provide a theoretical regulatory framework that aligns with the characteristics of the gaming industry in Macau and possibly other gaming jurisdictions.

As an important part of the global economy, family business have made a lot of contributions to the world economy. With the deepening of China's reform and opening up, the scale and volume of family business have increased dramatically in the national economy. In recent years, more and more family-owned enterprises have entered the list of “Top 500 Chinese Enterprises” and become an important part of the national economy. As a special enterprise organization, family business has certain characteristics. Most of the Chinese family business were founded in the early stage of China's reform and opening up, and now more and more family business are facing the problem that the founders need to leave the family business due to their advanced age. How to ensure the smooth succession of family business and avoid conflicts and contradictions in the process of succession is an important challenge faced by many family business. How to ensure that family business can continue to develop and maintain a high performance status is of great significance to the founders and their successors, as well as to the Chinese economy. Therefore, this research examines how the characteristics of successors affect enterprise performance based on the context of intergenerational succession in family business. This research is based on a multi-case study in which in-depth interviews were conducted with the founders and successors of five cases, which led to the conclusion that five successor characteristics have an impact on enterprise performance. They are inclusiveness, forward thinking, sociable and good communication, sense of social responsibility and sense of family mission. This finding has significant implications for the training of successors and the management of the succession process in Chinese family business. This study is conducive to the discovery of the mechanism of successor characteristics on enterprise performance as well as to make a realistic contribution to the research on the outcomes of family business in China.

Artificial Intelligence (AI) is being applied in different areas of Administration and Management including finance, e-commerce, etc. Project Management (PM) is one area that may benefit from the use of AI to support project managers in making more accurate predictions, more quickly, such as deadline adjustments and cost updates, while at the same time helping with some of repetitive tasks of PM by relieving managers from these processes. Nevertheless, multiple aspects are still in consideration to allow AI to be widely adopted in PM, including lack of validated systems, including aspects of quality and prevalence, trust from users, market and specialists, and how the government will play a role to support the wider adoption of AI tools. This research explores the integration of Artificial Intelligence (AI) in Project Management and its potential to enhance four aspects: service quality, trust, prevalence, and government support. The proposed methodology employs a systematic literature review (SLR) combining with a quantitative survey to assess the current state of AI in project management. The SLR covers scholarly articles from 2016 to 2021, focusing on AI's impact on project management across various industries. The survey, conducted among 200 professionals, gathers insights into AI's perceived benefits and challenges in project management. The research findings indicate a positive inclination towards AI in project management, with respondents recognizing its potential to improve efficiency, support data-driven decisions, and enhance risk management. However, the study also reveals concerns regarding data quality, privacy, and the need for ethical considerations in AI applications. Most respondents agree on the necessity of government support to foster AI adoption and the importance of establishing trust in AI systems through transparency and security measures. The thesis concludes with recommendations for practitioners and policymakers to effectively leverage AI in project management. It proposes a framework including the development of training programs, the establishment of quality standards for AI services, and the promotion of public-private partnerships to drive innovation. The study emphasizes the importance of a multi-faceted approach to AI integration, considering technological, organizational, and ethical dimensions.

The stock market's inherent volatility and complexity pose significant challenges for investors seeking to optimize their strategies. This thesis addresses the critical need for improved forecasting methods in stock price prediction by proposing a hybrid approach that combines traditional machine learning (ML) techniques, specifically Support Vector Machines (SVM) and Long Short-Term Memory (LSTM) networks, with sentiment analysis derived from financial news and social media platforms. The research establishes a theoretical framework integrating quantitative data, such as historical stock prices, with qualitative sentiment data to enhance prediction accuracy. The study involves the collection of a comprehensive dataset covering stock prices and sentiment scores from various sources, including news articles and social media posts, from January 2010 to December 2023. Rigorous data preprocessing techniques, including normalization and feature engineering, are employed to prepare the data for analysis. A comparative analysis of the SVM and LSTM models uses multiple performance metrics, including Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and classification accuracy. The findings reveal that the LSTM model significantly outperforms the SVM model in predictive accuracy, demonstrating its capability to capture complex temporal dependencies inherent in financial time series data. Furthermore, integrating sentiment analysis significantly enhances the predictive performance of both models. Notably, transformer-based sentiment analysis techniques, such as BERT and DistilBERT, provide superior sentiment classifications compared to traditional methods like VADER and TextBlob. The empirical results indicate that incorporating sentiment data leads to an average accuracy improvement of 12.8% over models that rely solely on historical price data. This research contributes to the evolving field of financial forecasting by emphasizing the importance of a hybrid approach that amalgamates quantitative and qualitative data. The implications of these findings extend beyond academic research, offering valuable insights for investors and financial analysts seeking to leverage advanced predictive models to navigate market uncertainties. Ultimately, this dissertation advocates adopting sophisticated hybrid models that enhance stock investment strategies and decision-making processes in the finance sector.

This study sought to determine the strategy that allowed the Las Vegas Sands Corporation (LVS) to attain its leading status in the casino industry and to gain insight whether this status would continue given (i) the passing of the LVS founder, Sheldon Adelson, in January 2021, (ii)the sell-off of the company's Las Vegas properties early in 2022, and (iii) the firm's greater sensitivity to events in China caused by the company's increased reliance for most of its customers on the mainland China market. The study first identified the nature of the LVS competitive advantages when Adelson was directing the firm and then assessed whether these had been adversely impacted due to changes in the firm's markets, management or strategy. The study relied initially on the work of David Baron, Professor of Political Economy and Strategy at Stanford University who as early as 1981 advanced the view that corporate strategy needed to be divided in a Marketing Strategy (MS) and a Non-Market Strategy (NMS). The NMS component for LVS was critically important since government determined who could acquire a Macau casino concession and what level of visas would be provided to Mainland China gamblers to fill the Macau casinos. The key question became the nature of Adelson's Political Effectiveness as determined by the NMS he directed towards the China market. To resolve this issue, we adopted the Wellner & Lakotta proposal to extend Porter's Five Forces analytical framework by two additional dimensions, Government Interventors and Complementor Organizations. We concluded that it was highly likely that Goldman Sachs, the long-term financial backer of Sheldon Adelson, played a significant if not the major role in the success Adelson was able to achieve in the Greater China market.