Search results

This paper develops a robust hospital readmission prediction framework by combining the feature selection algorithm and machine learning (ML) classifiers. The improved feature selection is proposed by considering the uncertainty in patient's attributes that leads to the output variable.

First, data preprocessing is conducted which includes how raw data is managed. Second, the impactful features are selected through feature selection process. It started with calculating the relational grade of each patient towards readmission using grey relational analysis (GRA) and the grade is used as the target values for feature selection. Then, the influenced features are selected using the Least Absolute Shrinkage and Selection Operator (LASSO) method. This proposed method is termed as Grey-LASSO feature selection. The final task is the readmission prediction using ML classifiers.

The proposed method offered good performances with a minimum feature subset up to 54–65% discarded features. Multi-Layer Perceptron with Grey-LASSO gave the best performance.

The performance of Grey-LASSO is justified in two readmission datasets. Further research is required to examine the generalisability to other datasets.

In designing the feature selection algorithm, the selection on influenced input variables was based on the integration of GRA and LASSO. Specifically, GRA is a part of the grey system theory, which was employed to analyse the relation between systems under uncertain conditions. The LASSO approach was adopted due to its ability for sparse data representation.

This paper aims to review the literature of community-based mental health services (CMHS) optimization in Malaysia that focus on the aspects of perceived challenges or barriers and facilitating factors for community-based mental health centre (CMHC).

Relevant peer-reviewed literature was identified and being reviewed in three databases and one local journal.

Nine studies were eligible for inclusion and the extracted data were summarized into a narrative review. The evidence provided for what constituting an optimal implementation of CMHS in Malaysia appeared lacking. However, there are adequate recommendations and suggestions on the component of facilitating factors; on how to increase service user’s engagement towards the improvement of mental health systems.

Research in the literature had focussed on the optimization aspects from the perspective of the service user. Additionally, this had been discussed through the lens of a variety of mental health systems and not specifically on the community-based perspectives.

This current systematic review provides valuable and important evidence that further explorations of the existing experiences of the mental health care professionals in delivering CMHS in Malaysia CMHC. The review serves as the foundation for future research to explain the local explanatory framework on what are the best practise models for local CMHS.

This paper reviews the extant research on Web3.0 published between 2003 and 2022.

This study uses a topic modeling procedure latent Dirichlet allocation to uncover the research themes and the key phrases associated with each theme.

This study uncovers seven research themes that have been featured in the existing research. In particular, the study highlights the interaction among the research themes that contribute to the understanding of a number of solutions, applications and use cases, such as metaverse and non-fungible tokens.

Despite the relatively small data size of the study, the results remain significant as they contribute to a more profound comprehension of the relevant field and offer guidance for future research directions. The previous analysis revealed that the current Web3.0 technology is still encountering several challenges. Building upon the pioneering research in the field of blockchain, decentralized networks, smart contracts and algorithms, the study proposes an exploratory agenda for future research from an ecosystem approach, targeting to enhance the current state of affairs.

Although topics around Web3.0 have been discussed intensively among the crypto community and technological enthusiasts, there is limited research that provides a comprehensive description of all the related issues and an in-depth analysis of their real-world implications from an ecosystem perspective.

The rapid rise of large language models (LLMs) has propelled them to the forefront of applications in natural language processing (NLP). This paper aims to present a comprehensive examination of the research landscape in LLMs, providing an overview of the prevailing themes and topics within this dynamic domain.

Drawing from an extensive corpus of 198 records published between 1996 to 2023 from the relevant academic database encompassing journal articles, books, book chapters, conference papers and selected working papers, this study delves deep into the multifaceted world of LLM research. In this study, the authors employed the BERTopic algorithm, a recent advancement in topic modeling, to conduct a comprehensive analysis of the data after it had been meticulously cleaned and preprocessed. BERTopic leverages the power of transformer-based language models like bidirectional encoder representations from transformers (BERT) to generate more meaningful and coherent topics. This approach facilitates the identification of hidden patterns within the data, enabling authors to uncover valuable insights that might otherwise have remained obscure. The analysis revealed four distinct clusters of topics in LLM research: “language and NLP”, “education and teaching”, “clinical and medical applications” and “speech and recognition techniques”. Each cluster embodies a unique aspect of LLM application and showcases the breadth of possibilities that LLM technology has to offer. In addition to presenting the research findings, this paper identifies key challenges and opportunities in the realm of LLMs. It underscores the necessity for further investigation in specific areas, including the paramount importance of addressing potential biases, transparency and explainability, data privacy and security, and responsible deployment of LLM technology.

The analysis revealed four distinct clusters of topics in LLM research: “language and NLP”, “education and teaching”, “clinical and medical applications” and “speech and recognition techniques”. Each cluster embodies a unique aspect of LLM application and showcases the breadth of possibilities that LLM technology has to offer. In addition to presenting the research findings, this paper identifies key challenges and opportunities in the realm of LLMs. It underscores the necessity for further investigation in specific areas, including the paramount importance of addressing potential biases, transparency and explainability, data privacy and security, and responsible deployment of LLM technology.

This classification offers practical guidance for researchers, developers, educators, and policymakers to focus efforts and resources. The study underscores the importance of addressing challenges in LLMs, including potential biases, transparency, data privacy, and responsible deployment. Policymakers can utilize this information to shape regulations, while developers can tailor technology development based on the diverse applications identified. The findings also emphasize the need for interdisciplinary collaboration and highlight ethical considerations, providing a roadmap for navigating the complex landscape of LLM research and applications.

This study stands out as the first to examine the evolution of LLMs across such a long time frame and across such diversified disciplines. It provides a unique perspective on the key areas of LLM research, highlighting the breadth and depth of LLM’s evolution.

This study systematically reviews the research landscape of humanitarian financing, identifying key themes and pathways to foster a resilient humanitarian value ecosystem.

A systematic review using BERTopic analysed 1,043 academic publications from the Web of Science database to uncover major themes in humanitarian financing literature.

Six core themes emerged from the study: Disaster Risk and Climate Change Resilience, Refugee Crises and Humanitarian Health Services, Preparedness and Response to Health Emergencies, Supply Chain and Logistics Fundraising, Food Security and Nutrition Insecurity and Global Health Challenges and Disease Control. The study highlights the role of financing in addressing short-term needs and promoting long-term systemic resilience, such as disaster preparedness, healthcare systems and food security.

This study provides a comprehensive overview of key research trends in humanitarian financing. However, the reliance on academic literature limits the inclusion of insights from grey literature and practitioner reports. Future studies should adopt broader data sources and interdisciplinary approaches.

Findings underscore the importance of integrating emergency funds with long-term resilience investments, enhancing financial transparency and advocating innovations to optimise humanitarian operations.

The study advocates for interdisciplinary approaches and policies connecting disaster risk reduction, healthcare, climate resilience and food security to build a sustainable humanitarian ecosystem.

This review proposes a novel conceptual framework based on a systematic review of literature using BERTopic, offering actionable insights into the role of humanitarian financing in addressing global crises and promoting resilience.

The purpose of this paper is to investigate political influential spreaders in Twitter at the juncture before and after the Malaysian General Election in 2013 (MGE2013) for the purpose of understanding if the political sphere within Twitter reflects the intentions, popularity and influence of political figures in the year in which Malaysia has its first “social media election.”

A Big Data approach was used for acquiring a series of longitudinal data sets during the election period. The work differs from existing methods focusing on the general statistics of the number of followers, supporters, sentiment analysis, etc. A retweeting network has been extracted from tweets and retweets and has been mapped to a novel information flow and propagation network we developed. The authors conducted quantitative studies using k-shell decomposition, which enables the construction of a quantitative Twitter political propagation sphere where members posited at the core areas are more influential than those in the outer circles and periphery.

The authors conducted a comparative study of the influential members of Twitter political propagation sphere on the election day and the day after. The authors found that representatives of political parties which are located at the center of the propagation network are winners of the presidential election. This may indicate that influential power within Twitter is positively related to the final election results, at least in MGE2013. Furthermore, a number of non-politicians located at the center of the propagation network also significantly influenced the election.

This research is based on a large electoral campaign in a specific election period, and within a predefined nation. While the result is significant and meaningful, more case studies are needed for generalized application for identifying potential winning candidates in future social-media fueled political elections.

The authors presented a simple yet effective model for identifying influential spreaders in the Twitter political sphere. The application of the authors’ approach yielded the conclusion that online “coreness” score has significant influence to the final offline electoral results. This presents great opportunities for applying the novel methodology in the upcoming Malaysian General Election in 2018. The discovery presented here can be used for understanding how different players of political parties engage themselves in the election game in Twitter. The approach can also be adopted as a factor of influence for offline electoral activities. The conception of a quantitative approach in electoral results greatly influenced by social media means that comparative studies could be made in future elections.

Existing works related to general elections of various nations have either bypassed or ignored the subtle links between online and offline influential propagations. The modeling of influence from social media using a longitudinal and multilayered approach is also rarely studied. This simple yet effective method provides a new perspective of practice for understanding how different players behave and mutually shape each other over time in the election game.

The purpose of this paper is to compare the curriculum developments of civic education in three emerging Chinese societies: China and two Special Administrative Regions of Hong Kong and Macao, which are increasingly under the impacts of globalisation in this information world.

The analytical method is used and the following are identified: active and global civic education-related learning units and key themes and main contents in official curriculum guidelines and updated textbooks related to civic education.

A major finding is that elements of both active and global citizenship, such as participation in the community and understanding about the world and thus forming multiple identities, can be found alongside their emphasis on enhancing national citizenship. Thus, ideas of global citizenship and multiple levels of citizenship from local, national to global start to develop in these three Chinese societies.

The implications of such findings of both active and global citizenship, as well as multiple identities, found in these three Chinese societies could be huge for informing civic literature and sociological point of views, in particular, pointing to the next generations receiving a broadened and transcended notion of multiple levels of citizenship, apart from local and national citizenship.

The significance of this paper is that it argues that ideas of active citizenship in terms of community participation and global citizenship have been found in China, Hong Kong and Macao civic education curriculum and textbooks because of the expectations placed on students to compete in a globalized world, though national citizenship and patriotic concerns have been primary concerns. Globalisation makes the world society by impacting on these three Chinese societies for active and global citizenship, though they still retain their particular curricular focusses.

This paper aims to explore factors affecting young consumers’ purchasing of new smart products under the influence of social media.

Twenty semi-structured, face-to-face, in-depth interviews were conducted with young consumers who used new media, and five extended interviews were conducted with popular smartwatch retailers in Macau.

The findings reveal that they tend to garner product information from multiple channels of communication. Social media exert the greatest influence. Reliable information, strong branding and interactions with vendors are also influential, although new product pre-announcements may be boring and difficult to understand.

This study presents new insights into diffusion of innovation theory and provides retailers launching smart products with a better understanding of their target young customers’ purchasing behavior.

The purpose of this paper is to identify the continuance usage of livestreaming commerce. The proposed antecedents of livestreaming commerce continuance usage are social presence, perceived crowdedness, need for arousal, susceptibility to informational influence, and perceived value, whilst trust in broadcasters is treated as the moderating factor.

A survey was conducted using a questionnaire designed specifically for this purpose, which was completed by 388 Malaysian consumers aged 26–40 years who have prior experience with livestreaming commerce. The authors tested the framework using the partial least squares structural equation modelling (PLS-SEM) technique and complementary by performing artificial neural network (ANN) analysis.

The results indicate several significant relationships, whereby social presence, perceived crowdedness, susceptibility to informational influence, and trust in broadcasters significantly influence perceived value, whereas perceived value significantly influences continuance intention. In addition, trust in broadcasters moderates the relationship between perceived value and continuance intention.

The current study adds significantly to the body of knowledge about livestreaming by examining the direct determinants of perceived value and thus delineating how perceived value influences continuance intention. The findings yielded also illustrate the role of trust in broadcaster as the moderator of the link between perceived value and continuance intention.

The study aims to examine the relationships among personality traits (i.e. the Big Five personality traits and locus of control), self-perceived facial attractiveness, motivations (i.e. intrinsic and extrinsic motivation) and intention toward live stream broadcasting. It also investigates the moderating role of perceived behavioral control in the relationship between motivations and intention.

Data collected from a sample of 637 participants are used to examine the research model and test the hypotheses with the employment of partial least squares structural equation modeling.

The study shows that motivations and perceived behavioral control are significant predictors of intention. Perceived behavioral control has a significant moderating effect between motivations and intention. Intrinsic motivation is positively influenced by self-perceived facial attractiveness, agreeableness, extraversion and internal locus of control, while extrinsic motivation is positively predicted by self-perceived facial attractiveness, conscientiousness and extraversion.

This study enhances our understanding of the determinants of intention toward live stream broadcasting by exploring its relationships with motivations, self-perceived facial attractiveness and personality, as well as the moderating effects of perceived behavioral control.