39506 research outputs found
Sort by
AMPHIBIANS
No full textThe chapter enumerates the amphibian fauna of Niah National Park, Miri, Sarawak
A lightweight neural attention-based model for service chatbots
No full textThe growing demand for efficient service chatbots has led to the development of various deep learning techniques, such as generative neural attention-based mechanisms. However, existing attention processes often face challenges in generating contextually relevant responses. This study introduces a lightweight neural attention mechanism designed to enhance the scalability of service chatbots by integrating a scalar function into the existing attention score computation. While inspired by scaling practices in transformer models, the proposed scalar is tailored to seq2seq architectures to optimize the alignment sequences, resulting in improved context relevance and reduced resource requirements. To validate its effectiveness, the proposed model was evaluated on a real-world Customer Support Twitter dataset. Experimental results demonstrate a +0.82 BLEU-4 improvement and a 28% reduction in training time per epoch over the baseline. Moreover, the model achieves the target validation loss two epochs earlier, indicating faster convergence and improved training stability. Further experiments investigated activation functions and weight initializers integrated into the proposed model to identify optimal configurations that optimize the model’s performance. Comparative experimental results show that the proposed modifications significantly enhance response accuracy and contextual relevance. This lightweight attention mechanism addresses key limitations of existing attention mechanisms. Future work may extend this approach by combining it with transformer-based architectures to support broader sequence prediction tasks, including machine translation, recommender systems, and image captioning
Transition of psychological state from recording to live performance : a longitudinal explanatory sequential study in music students from Universiti Malaya during COVID-19
No full textMusic students often struggle with music performance anxiety (MPA), which is debilitating stress brought on by performances. However, MPA varies according to the type of performance. Therefore, this study aims to compare the
MPA levels of performance recorded versus live. This study also looks at the factors that influence students’ anxiety levels and their views on the various assessment formats. A total of 28 second-year music undergraduates were
recruited. Survey questionnaires (K-MPAI-R) were administered to evaluate the anxiety levels of the students. This was followed by interview sessions with randomly selected participants. Through descriptive statistics and paired sample t-tests, this study suggests that the participants were afflicted with different levels of MPA in the first year (M = 127.379) compared to the second year
(M = 138.621). There was a significant difference between the MPA level in the recorded assessment and the live assessment. The qualitative interview data were analysed through thematic analysis. The following themes emerged
to reflect the factors that influence MPA. Mental/psychological stress was the emergent theme for internal factors, whereas environment, sound quality, and
physical challenges were emergent themes for external factors. In conclusion, MPA levels of the students were high regardless of the assessment format and the factors that resulted in MPA varied according to the format
Synthesis and Application of Graphene Oxide@ZIF-8/Alumina Hollow Fibre Membrane for Oily Wastewater Treatment
No full textOily wastewater pollution primarily harms drinking water and groundwater resources, endangers marine life and threatens human health. Therefore, the use of membrane filtration technology for water reclamation from oily wastewater is one of the effective ways for the removal of oil from industrial wastewater. Hence a promising approach is by microfiltration using membrane-based graphene oxide (GO). The pristine alumina hollow fibre membrane is modified with a graphene oxide (GO) sheet which was inserted with Zeolitic Imidazolate Framework-8 (ZIF-8) crystals to form a stable nanocomposite membrane with a high surface area and rich with oxygen functional groups. This significantly improved the adsorption capacity because it provides more site for oil adsorption. Therefore, this research aims to synthesize the graphene oxide (GO) and to fabricate the GO@ZIF-8 nanocomposites onto the alumina hollow fibre membrane (Al2O3) to produced hybrid graphene oxide@ZIF-8/alumina hollow fibre membrane (GO@ZIF-8/Al2O3). In brief, GO is prepared by the oxidation of graphite using KMnO4 as a strong oxidizing agent based on the modified Hummers’ method, then followed by the fabrication of crosslinked GO@ZIF-8 composites on the alumina hollow fibre membrane via in-situ solvothermal assembly method. The GO@ZIF-8/Al2O3 composites structure is explored by Fourier-Transform Infrared Spectrophotometer (FTIR), Ultraviolet–visible spectroscopy (UV-Vis), Field Emission Scanning Electron Microscope (FE-SEM) and X-Ray Diffraction (XRD). Successively, the removal of oil from water by GO@ZIF-8/Al2O3 membrane performance was evaluated through the batch adsorption study and crossflow filtration system. The effects of several adsorption factors, such as initial oil concentration and response time, were investigated. To determine the adsorption mechanism, different kinetic models including Pseudo-First Order and Pseudo-Second Order and isotherms models such as Freundlich and Langmuir were used. Adsorption isotherm is best suited by the Freundlich model (R2=0.98046), whereas adsorption kinetics is best fit by the pseudo-second order (R2=0.98247), indicating that the adsorption process involves chemisorption. According to the crossflow filtration study, the membranes could remove up to 96.32% of the oil from the aqueous solution. Additionally, the reusability analysis revealed that 90.08% of the oil was rejected after 6 reusability cycles. The better separation performance could be attributed to the well dispersion of ZIF-8 nanoparticles in the GO onto the alumina hollow fibre support membrane. Hence, a straightforward scalable approach for producing reliable GO@ZIF-8 based membranes with an efficient oil removal capability is shown
ESTIMATING CARBON STOCKS IN Shorea macrophylla RESTORATION SITES, SARAWAK
No full textEstimating carbon stocks in forest landscape restoration ( sites provides valuable information into their potential role in carbon sequestration and ecosystem recovery This study quantified soil organic carbon ( and total tree carbon stocks (aboveground and belowground) in enrichment planting sites of Shorea macrophylla (de Vriese) P S Ashton in Gunung Apeng National Park, Sarawak Soil samples 0 30 cm depth) were analyzed for bulk density, organic matter and total carbon, while stand level tree measurements at different age stands S. macrophylla planted after 4 6 12 and 15 years) were used to estimate biomass carbon using established allometric approaches Results showed SOC in the topsoil ranged from 512 72 to 1700 84 g m⁻², with higher values in older stands, suggesting gradual improvement of soil
conditions through litter input and canopy development Total tree carbon stocks ranged between 15 03 and 506 60 t ha⁻¹, reflecting variation in tree growth and survival rates among
stands The focus on S macrophylla highlights the long term sequestration potential of dipterocarp species in forest restoration sites Overall, these findings provide baseline data for carbon stock estimation in Sarawak’s FLR initiatives and support ongoing monitoring and planning for climate mitigation strategie
Formulation of the SVEIR Model for COVID-19 by incorporating Control Intervention in Pakistan
No full textLike other countries around the world, Pakistan experienced a COVID-19 outbreak, starting from February 26, 2020, when the first two cases of COVID-19 were reported. To save the nation from this disease, the government of Pakistan took different strict measures. Along with the various aspects of the disease, including its epidemiology, transmission, clinical features, diagnosis, treatment, and prevention, it is widely accepted that mathematical models can be helpful to predict the occurrence of infectious diseases. However, the control interventions incorporating the Non-pharmaceutical interventions (NPI) measure for the current scenario in Pakistan, which include vaccination compartments, have not yet been identified for COVID-19. Therefore, the main objective of this study was to propose a new COVID-19 model by incorporating the vaccine factor to analyze the COVID-19 trends in Pakistan. In order to achieve the objective of research, a systematic process was followed, and an analogous existing model called STEQIR (Gill et al., 2020) was evaluated. Then, by using some modifications such as transitioning individuals from the susceptible population to a vaccinated population and by compartmental analysis, including the control intervention of the vaccine, a new model, namely the SVEIR (Susceptible, Vaccinated, Exposed, Infected, and Recovered) model, was derived. The proposed model is based on the assumption that Vaccination reduces the effective susceptible group, lowering the force of infection. The possibility of the analytical solution of the proposed SVEIR model was investigated. However, the analytical solution for the SVEIR model is quite tricky and impractical due to high non-linearity and the implicit nature of the governing equations. Therefore, the numerical solution of the proposed SVEIR model was computed using a higher-order Runge-Kutta method, which yields a stable solution of the SVEIR model. The numerical simulation of all five variables can project the outbreak well over the next few months. It was established that despite the large population of Pakistan, the measured maximum percentage error (MAPE) was in an acceptable range for all five variables that fell between 9.72 and 15.42, and the results were statistically significant at 90% and 95% confidence levels. The validation of the proposed SVEIR model was also done by comparing a benchmarking method (Gill, 2020), and it was shown that the SVEIR model performs better for the Pakistan COVID-19 data. To assess the variations caused by the working parameters involved in the SVEIR model, a sensitivity analysis was conducted, and the impact of all parameters involved in the SVEIR model was analyzed. From the analysis, it was revealed that among all the parameters, the transmission rate, death rate ε, and the average number of contacts per day per case are sensitive as they produce more variations in almost all population classes. Finally, the simulation of the basic reproduction number R0 regarding sensitive parameters and values was obtained. It was pointed out that the R0 remains more dominant for the probability of susceptibility β, becoming infectious per contact, and death rate; for other parameters, it remains less than 1. The outcome and findings of this study may be essential for policymakers and public health officials in making informed decisions and formulating effective strategies to govern and mitigate the spread of COVID-19. The understanding gained from this research may guide targeted strategies, promote efficient resource allocation, and contribute to the continuing efforts to mitigate and control the COVID-19 pandemic in Pakistan. Future research may focus on extending the proposed SVEIR model by incorporating additional factors, such as the efficacy of various vaccination techniques, the effects of non-pharmaceutical therapies, demographic characteristics, economic determinants, and seasonal variations in environmental and weather conditions weather
Advancing Academic Librarians’ Professional Development in Information Literacy Instruction
No full textThe expansion of higher education in the digital era has increased the demand for effective Information Literacy (IL) instruction and underscored the role of academic librarians. This study employed an exploratory sequential mixed-methods design to examine how Malaysian academic librarians implement theoretical frameworks in their teaching and to assess the effectiveness of a purpose-built IL MOOC in developing student competencies. Interview data indicated that IL instruction was informed by theoretical foundations through differentiated and scaffolded methods, ethical integration, interactive pedagogy, and continuous professional development. Librarians utilized the TPACK framework by integrating digital tools with learner-centered strategies, while the ACRL Framework informed the development of structured modules, practical search and evaluation activities, reciprocal engagement, and assessment practices. Quantitative results indicated that time invested in the MOOC was the most significant predictor of course completion, while participation in activities and repeated assessments further improved outcomes. Engagement and course design had a greater impact than video duration. Competency analysis demonstrated significant differences between completers and non-completers across all six IL modules, with the largest improvements observed in 'Scholarship as Conversation' and 'Searching as Strategic Exploration.' These findings underscore the importance of aligning IL instruction with the TPACK and ACRL frameworks, enhancing librarians’ pedagogical expertise, and utilizing MOOCs as scalable solutions for improving IL competencies
Penerokaan Pengajaran Bahasa Melayu Komunikasi melalui Pengintegrasian PTPK-UDL untuk Murid Berkeperluan Pendidikan Khas Kurang Upaya Intelektual di Program Pendidikan Khas Integrasi
No full textKajian ini bertujuan meneroka pengintegrasian Pengetahuan Teknologi, Pedagogi dan Kandungan (PTPK) serta Reka Bentuk Pembelajaran Sejagat (UDL) dalam pengajaran Bahasa Melayu Komunikasi (BMK) untuk Murid Berkeperluan Pendidikan Khas Kurang Upaya Intelektual (MBPK KUI) di Program Pendidikan Khas Integrasi (PPKI). Pengajaran BMK untuk MBPK KUI di PPKI memerlukan pendekatan yang khusus melalui penguasaan pelbagai pengetahuan dan kemahiran guru. Namun, sejauh mana pengetahuan teknologi, pedagogi, kandungan dan prinsip UDL dalam kalangan guru PPKI masih perlu diterokai untuk memastikan pengajaran yang berkesan. Pengintegrasian kesemua elemen ini dalam pengajaran BMK untuk MBPK KUI juga masih kurang diberi perhatian dalam kajian sedia ada, terutamanya di PPKI. Kajian kualitatif ini melibatkan empat orang guru PPKI yang mengajar BMK di dua buah sekolah berlainan. Data dikumpul melalui temu bual separa berstruktur, pemerhatian di dalam kelas, dan analisis dokumen. Dapatan kajian menunjukkan penggunaan seimbang antara teknologi rendah dan tinggi dalam pengajaran BMK dengan pengintegrasian prinsip UDL yang menyokong kepelbagaian cara pembelajaran. Strategi pedagogi yang pelbagai dan fleksibel digunakan untuk memenuhi keperluan pembelajaran MBPK KUI yang berbeza, manakala kandungan BMK disesuaikan mengikut tahap keupayaan murid melalui penggunaan situasi autentik. Amalan terbaik yang dikenal pasti termasuk program mentor BMK, simulasi situasi harian, penghasilan bahan pembelajaran digital, dan carta pemantauan visual. Dapatan kajian mendapati, Model PTPK-UDL Benton-Borghi telah diadaptasi untuk menghasilkan satu kerangka pengajaran BMK yang komprehensif dan sesuai dengan konteks di PPKI. Model yang diadaptasi ini menggabungkan elemen PTPK dan prinsip UDL dengan berkesan dan disokong oleh teori-teori pembelajaran yang relevan bagi menyediakan satu kerangka pengajaran yang mengambil kira keperluan khusus dan realiti pelaksanaan pengajaran BMK untuk MBPK KUI di PPKI
Organic-Based fertilizers Produced from Sago Hampas and Decanter Cake and Their Uses in Maize (Zea Mays L.) Cultivation
No full textSustainable agriculture requires innovative strategies to enhance soil fertility while reducing the environmental impacts of agro-industrial waste. This study investigated the synergistic effects of Black Soldier Fly (BSF) frass and decanter cake (DC) biochar on maize (Zea Mays L.) growth and soil fertility in tropical acidic soil. BSF larvae were used to bioconvert sago hampas and DC into nutrient-rich frass, while DC was pyrolyzed to produce biochar. Laboratory incubation and pot experiments were conducted with eight treatments: control (F0B0C0), 100% chemical fertilizer (F0B0C1), 100% BSF frass alone (F1B0C0), 100% biochar alone (F0B1C0), 50% frass + biochar + chemical fertilizer (F2B2C2), 50% frass + chemical fertilizer (F2B0C2), 50% biochar + chemical fertilizer (F0B2C2), and 50% frass + biochar (F2B2C0). Soil chemical properties, nutrient availability, and maize biomass were assessed after 52 days. Results showed that the combined application of frass and biochar significantly improved soil pH, cation exchange capacity, and nutrient retention compared to untreated soil. The integrated treatment (F2B2C2) recorded the highest improvements, with a 41% increase in total nitrogen, 58% enhancement in phosphorus, 22% improvement in potassium, and a 37% increase in zinc relative to the control. These nutrient enhancements translated into the highest maize dry weight among treatments. The study concludes that BSF frass and DC biochar, particularly in combination with reduced chemical fertilizers, enhance nutrient-use efficiency, sustain soil fertility, and improve maize productivity. This integrated approach valorizes agro-industrial wastes, reduces dependence on synthetic inputs, and contributes to sustainable tropical agriculture
Investigating the Phylogenetic Diversity of SARS-CoV-Like Viruses in Various Species of Bats in Sarawak, Malaysia
No full textBats are natural reservoirs for a diverse range of coronaviruses (CoVs), including those closely related to SARS-CoV and SARS-CoV-2. Their unique ecological traits, including high mobility, diverse dietary habits, and social roosting behavior, facilitate viral maintenance and transmission, making them critical subjects for understanding zoonotic spillover risks. Despite Malaysia's rich bat diversity, particularly in Sarawak, studies on bat CoVs (BtCoVs) remain limited. This study aims to fill this gap by investigating the prevalence, diversity, and phylogenetic relationships of bat CoVs in western and northern Sarawak, Malaysian Borneo. A total of 346 fecal samples were collected from 29 bat species spanning six families, including Pteropodidae, Hipposideridae, Vespertilionidae, Miniopteridae, Nycteridae, and Rhinolophidae. To ensure comprehensive detection, two widely used molecular assays, the Quan (Q-assay) and Watanabe (W-assay) pan-CoV PCR protocols, were employed. The selection of these assays was based on their established sensitivity in detecting both known and potentially novel CoV lineages by targeting conserved regions of the RNA-dependent RNA polymerase (RdRp) gene. The Q-assay has been widely utilized for detecting a broad range of CoVs across various host species, while the W-assay is optimized for bat-associated CoVs, making their combined application suitable for evaluating detection efficacy. The performance of both assays was statistically compared using the Kappa statistic and McNemar’s test to assess agreement and detection differences. The results revealed a CoV prevalence of 14.45% with the Q-assay and 12.72% with the W-assay, with a combined prevalence of 22.83%. The two assays demonstrated fair agreement (κ=0.286, p0.05), indicating comparable performance in identifying bat CoVs. Phylogenetic analysis of RdRp sequences identified six distinct clades within alphacoronaviruses (α-CoVs) and betacoronaviruses (β-CoVs), including two unclassified Borneo-α-CoVs and four classified CoVs within the subgenera Minunacovirus, Rhinacovirus, Nobecovirus, and Sarbecovirus. Notably, Sarbecovirus-related sequences were detected in Rhinolophidae bats, further supporting their role as key reservoirs of viruses related to SARS-CoV and SARS-CoV-2. The use of two molecular assays provided a more comprehensive picture of CoV diversity and demonstrated the importance of using complementary detection methods. While both assays showed similar detection capabilities, their combined use allowed for the identification of a broader range of CoVs, including potentially novel strains. This study represents the first detailed phylogenetic analysis of bat CoVs in Sarawak derived from fecal samples, addressing a significant research gap. The findings underscore the importance of continued surveillance efforts in high-biodiversity regions to improve understanding of CoV ecology and mitigate zoonotic risks. Future research should explore whole-genome sequencing approaches to further characterize unclassified CoV clades and refine molecular assays for enhanced detection.
Keywords: Bat coronaviruses, Q-assay, W-assay, Sarbecoviruses, zoonoti