39506 research outputs found
Sort by
THE INFLUENCE OF FINANCIAL RISK ON BANK STOCK RETURN IN MALAYSIA
No full textThis study investigates how financial risk influences the stock returns of Malaysian banks listed
on Bursa Malaysia from 2015 to 2024. Using a multivariate Generalized Least Squares
regression model, the effects of credit risk, market risk, liquidity risk, and capital risk on
quarterly bank stock performance were examined. The analysis leverages data from nine major
listed banks, controlling for multicollinearity, heteroscedasticity, and autocorrelation. The
findings reveal that among the four risk measures, only the capital-to-asset ratio has a
statistically significant positive impact on stock returns, indicating that stronger capitalization
enhances investor confidence. Credit, market, and liquidity risks do not show significant effects,
suggesting market efficiency in incorporating public information. These results offer insights
for investors, bank management, and regulators to improve risk mitigation strategies and
bolster financial stability in Malaysia's banking sector
First in class: The earliest appearance of major groups of amphibians and reptiles in postage stamps of the world
No full textThe first postage stamps to depict extant herpetological taxa, including members of the Anura, Caudata, Gymnophiona, Cro- codylia, Testudines, and Squamata (including Rhynchocephalia, Sauria, and Squamata), and familiar families within these orders, are discussed. Initial releases tend to be part of what was then re- ferred to as ‘pictorial series’, and show topics such as monarchs and other rulers, landscapes and other nationalistic topics, while more contemporary ones are thematic, drawing attention to the faunal groups
The Role of Social Support in the Face of Vulnerability to Economic Strain: Perspective from Sarawak Urban Area
No full textThis research investigates the complex interactions between economic strain, social support, resilience, and well-being in Sarawak in the aftermath of the economic disruptions brought about by the COVID-19 pandemic. Based on a sample of 290 participants, the study utilized structural equation modeling to address three primary objectives: examining the relationships between economic strain, social support, well-being, and resilience; assessing the mediating role of social support in the relationship between economic strain and well-being; and evaluating its mediation effect between economic strain and resilience. The initial findings corroborate the presence of significant associations between economic strain and the three constructs of well-being, resilience, and social support. Social support has been identified as a crucial mediator that modulates the interaction between economic strain and well-being and between economic strain and resilience. This abstract succinctly summarizes the central inquiry of the study, focusing on how economic adversities in the post-pandemic period influenced individual and community well-being in Sarawak, with an emphasis on the mitigating role of social support networks
Enhanced Model Compression for Lipreading Recognition based on Knowledge Distillation Algorithm
No full textLipreading is understanding what a speaker is saying by observing changes in the speaker's mouth. The lipreading recognition model LipPC-Net proposed in this
paper is built with a large Chinese lipreading data set based on Chinese phonetic rules and grammatical features and consists of two main parts: the P2P sub-model
and the P2C sub-model. The P2P sub-model is a model for identifying pinyin sequences from pictures, while the P2C sub-model is a model for identifying Chinese character sequences from pinyin. However, Chinese language features are rich and fuzzy, and the training optimization of lip-reading model requires high GPU computation and storage, so it is difficult to realize large-scale application.
Therefore, three knowledge distillation compression algorithms are proposed in this paper: Three different knowledge distillation compression algorithms, an
offline model compression algorithm based on multi-feature transfer (MTOF), an online model compression algorithm based on adversarial learning (ALON), and an online model compression algorithm based on consistent regularization(CRON) to complete the compression of the Chinese character sequence output by the model.
Three compression algorithms are used to fit and learn the transformation between different features, so that portable mobile terminals with limited hardware resources can carry the model. Thus, it can realize the practical application value of assisting the communication of deaf-mutes
The effect of bingo-KDK game to improve preservice counselor communication skills
No full textBasic communication skills are primary modalities for the counselor and should be developed from their preservice education. In Indonesia, most guidance and counseling major provide the basic communication skills course as a hybrid course, containing both theoretical and practical learning. This model had many challenges in making a balanced outcome for both the theoretical knowledge and practical skills. This research aims to test the bingo-keterampilan dasar komunikasi (KDK) as an integrated learning strategy of bingo games, reverse role-play, and reflection sessions to improve
preservice counselors’ basic communication skills. This research used a quantitative method with a randomized control trial (RCT) experiment design. This research involved 56 preservice counselors, divided randomly into
experiment and control groups. The measurement used is the performance test of the basic communication skills course. Data analysis focused on measuring basic statistics and t-tests. The results show the experiment group has higher
results and is significantly different from the control group. This effectiveness was supported by the learning atmosphere that indirectly increased the involvement of the preservice counselor. This research suggests exploring the
mental experience and performance measurement during the learning sessions using the bingo-KDK game
Improving Support Vector Machine Performance using Modified Similarity Distance Plotting-Data Reduction
No full textThe Support Vector Machine (SVM) is well-regarded for its high classification accuracy, but its computational efficiency is often challenged by large datasets. To address this, we introduce the Similarity Distance Plotting-Data Reduction (SDP-DR) method, a novel instance selection technique aimed at enhancing SVM's efficiency and generalization. SDP-DR utilizes similarity distances within the dataset to reduce the number of training instances, thereby lowering both the time and space complexities of the SVM algorithm. With a linear time complexity, which scales proportionally with the size of the dataset, SDP-DR is well-suited for large-scale datasets as it ensures faster processing times and lower computational costs. Our evaluation, conducted on 30 diverse datasets, compares the performance of SDP-DR with standard SVM, Edited Nearest Neighbour (ENN) + SVM, and Condensed Nearest Neighbour (CNN) + SVM models. The results reveal that SDP-DR, especially its Mean of Each Column (MEC) variant, achieves higher accuracy and competitive reduction rates while maintaining reasonable classification times compared to other benchmark algorithms. This balance positions SDP-DR as a promising approach for improving SVM performance, particularly in resource-constrained environments and large-scale datasets. By effectively reducing training instances, SDP-DR offers a pathway to more efficient machine learning models, making it a valuable contribution to instance selection research and advancing the development of scalable SVM classifiers. Its potential applications extend to large-scale data analysis, big data environments, and real-time machine learning systems where computational efficiency is critical
The Bidayuh Language of Sarawak, Malaysia: Language Use and Proficiency.
No full textBidayuh is an indigenous group living in Sarawak, Malaysia. The Bidayuh language is currently not taught in schools but experimented as a medium of instruction in some kindergartens. Past studies have shown that factors such as rural– urban migration, intermarriage, and higher education are causing a shift away from the Bidayuh language. However, they are conducted on a small-scale and little is known about the language maintenance situation among its community. This chapter presents the findings of Bidayuh language use and proficiency in the community. Analysis of the questionnaire data from 467 Bidayuh respondents showed that only 72.2% spoke Bidayuh as their first language, indicating that many no longer learn/ speak it as first language. On a daily basis, 66.6% reported speaking it. Based on their parentage, 71.7% considered themselves as pure Bidayuh, 27.84% as half-Bidayuh, and 0.43% felt that they had lost most of their Bidayuh characteristics. The findings showed that the respondents have better ability in understanding a conversation in the Bidayuh language compared to speaking it. They have lower literacy in the Bidayuh language because it is a spoken language and thus, not many learn it through formal instruction. Additionally, the large regional variations in the Bidayuh varieties accentuate the problems of standardizing the Bidayuh language for instruction in schools. A comparison of the results for the youngest and oldest groups of respondents indicate that the Bidayuh language maintenance is difficult in the face of the importance of English for career, and perceptions of Sarawak Malay dialect, English and Iban as strong languages in Sarawak.
Keywords Bidayuh · Language maintenance · Language use · Language proficiency · Identity · Sarawa
The influence of factors related to public health campaigns on vaccination behavior among population of Wuxi region, China
No full textBackground: Public health campaigns are essential for promoting vaccination behavior, but factors such as socioeconomic status, geographical location,
campaign quality, and service accessibility influence vaccine uptake. In the Wuxi region of China, disparities in vaccination behavior are seen between urban and
rural populations and among different socioeconomic groups. This study aims to explore the factors related to public health campaigns that affect vaccination behavior in Wuxi, contributing to better public health strategies.
Methods: A cross-sectional survey was conducted among 750 participants in Wuxi, focusing on their perceptions of socioeconomic status, geographical location, health campaign quality, and vaccination convenience. The questionnaire was developed based on a literature review and expert input using the Delphi method. Data were analyzed using descriptive statistics, reliability and validity tests, correlation analysis, and regression analysis, employing both SPSS and R software. Results: Socioeconomic status, geographic location, campaign quality, and accessibility all significantly influence vaccination behavior. Higher
socioeconomic backgrounds, urban residency, better campaign quality, and greater accessibility to vaccination services are positively correlated with higher vaccination uptake. Regression analysis revealed that public health campaigns
and accessibility are particularly influential in promoting vaccination behavior.
Conclusion: To improve vaccination rates, targeted strategies focusing on low socioeconomic groups, rural areas, and improving campaign quality and service
accessibility are necessary. Public health campaigns should be clear, culturally relevant, and utilize multiple communication channels. Future research should address misinformation, explore behavioral economics, and integrate emerging technologies like AI to optimize vaccination efforts
EDUCATION AND INEQUALITY IN THE ERA OF GLOBALISATION: COMPARATIVE POLICY INSIGHTS FROM TEN NATIONS
No full textGlobalisation has transformed national economies, social structures, and labour markets, presenting new opportunities but exacerbating income inequality and resource accessibility. Education is important to this shift, both as a conduit for social mobility and a factor in economic resilience. This study examines the relationship among globalisation, educational access, and economic inequality through a comparative analysis of 10 nations: the United States, Germany, Poland, Malaysia, Thailand, India, Indonesia, Bangladesh, Brazil, and Georgia. The study integrates secondary macroeconomic data from 2022 to 2023 with qualitative policy analyses. Statistical analysis employing Pearson correlation indicates minimal correlations among Gross Domestic Product (GDP) growth, secondary school enrolment, and income inequality, implying that merely increasing access to education is insufficient for achieving equitable development. Case study findings highlight that aligning education policy with labour market demands, promoting digital inclusion, and implementing decentralised governance structures substantially improves the efficacy of educational investments. India and Poland exemplify effective methods through vocational integration and equity-centered reforms, but enduring disparities in Brazil and Malaysia underscore the shortcomings of access-oriented initiatives. This study highlights the necessity of focussing on education within an integrated framework that encompasses social protection, economic inclusion, and sustainable policy development. It provides evidence-based suggestions consistent with Sustainable Development Goals 4 (Quality Education) and 10 (Reduced Inequalities), promoting inclusive educational systems that effectively address structural disparities exacerbated by globalisation
What Drives Satisfaction in Fresh E-Commerce? Evidence from Review-Based Topic and Sentiment Analysis
No full textThis study explores key determinants of customer satisfaction in China’s fresh e-commerce sector by analyzing large-scale user-generated reviews from JD Fresh. Unlike prior research relying on surveys, this study integrates semantic segmentation, high-frequency textual data, and multivariate modeling to offer a data-driven and fine-grained understanding of customer experience. Nine service dimensions were identified: Freshness, Affordability, Size, Taste, Ice Pack, Delivery, Quality, Storage, and Platform Functionality. Sentiment scores were calculated for each dimension and used in a multiple linear regression model to assess their impact on overall satisfaction. Results indicate that all nine factors significantly affect satisfaction, with Taste, Quality, and Storage being the most influential. These findings demonstrate the effectiveness of combining text mining and sentiment analysis in identifying service quality dimensions and predicting satisfaction. The study also highlights the value of online reviews as a scalable source for service evaluation and provides actionable insights for improving user experience and repurchase intention in the highly competitive fresh e-commerce industry