39506 research outputs found
Sort by
Exploring the Effect of Zeolite/ Activated Carbon Variation in Filtration Media for Peat Water Treatmen
No full textDirect consumption of untreated peat water poses health risks due to the presence of humic substances. Hence,
this study examines the influence of different ratios of zeolite to activated carbon on removing contaminants
from peat water, utilizing a fixed amount of zeolite (30 g) with varying activated carbon weight (0, 10, 20, 30 g).
The highest humic content removal was by column D (30 g zeolite: 30 g activated carbon; 1:1 ratio) at 66.67 %,
and similar findings for chemical oxygen demand (COD) and total organic carbon (TOC), single column named
as column D remove these water quality indicators at 84.45 % and 65.95 %, respectively. The adsorption of
humic acid on carbon's surface is attributed to a combination of electrostatic attraction and surface complex
formation
Intelligent Threat Determination and PM in Unmanned Aerial Vehicles: A Review of Deep Learning Perspectives
No full textThis study delves into the impact of Deep Learning (DL) on UAV advancements from 2015 to 2025, focusing on early threat detection, anomaly recognition, and Predictive Maintenance (PM). By reviewing state-of-the-art DL models, we assess their role in mitigating cybersecurity threats, detecting flight anomalies, and optimizing UAV reliability. Public datasets such as ALFA and UAV Attack Dataset have been instrumental in evaluating these models for real-world applications. CNNs excel in spatial threat detection, LSTMs in sequential anomaly recognition, and Transformers in multi-modal sensor fusion. Despite these advancements, computational constraints, adversarial vulnerabilities, and real-time processing challenges persist. Future research must focus on energy-efficient AI, explainable models, and swarm intelligence to enhance UAV autonomy. This review provides a comprehensive decade-long perspective on DL-driven UAV innovations, shaping future developments in security and PM
Successful Ageing and Its Associated Factors Among Elderly Association Members in Kuching
No full textBackground: Successful ageing has become vital in the growing elderly population. The current study aimed to determine the prevalence and factors linked with successful ageing among members of elderly associations in Kuching.
Materials and methods: This study employed a structured questionnaire from the Successful Ageing Inventory (SAI)
through a cross-sectional survey involving 172 respondents. Respondents were randomly sampled from elderly associations, and information were collected from January to April 2024. Subsequently, statistical package for social sciences (SPSS) 22.0 was utilised during data analysis, employing descriptive statistics, univariate analysis, and multiple
linear regression. Results: A 66.28% successful ageing prevalence was recorded. The results also indicated positive
associations with successful ageing with higher education (secondary education: Adj. b = .85, 95% CI: .04, 1.66; p
= .040; tertiary education or higher: Adj. b = 0.97, 95% CI: .09, 1.85; p = .031), living arrangements (spouse: Adj. b
= 1.24, 95% CI: 0.38, 2.10; p = .005; spouse and children: Adj. b = 1.10, 95% CI: 0.23, 1.97; p = .014), Similarly,
respondents who had five or more close friends or neighbors (Adj. b = 1.41, 95% CI: 0.19, 2.62; p = .023), and better access to neighborhood facilities (Adj. b = 1.49, 95% CI: 0.19, 2.79; p =.025) demonstrated successful ageing.
Conversely, heavy drinking negatively impacted successful ageing among the respondents (Adj. b = -1.44, 95% CI:
-2.65, -0.22; p = .021). Conclusion: Based on the findings, primary parameters influencing successful ageing include
drinking habits, education level, living arrangements, access to neighbourhood facilities and social support through participation in elderly association activities
A systematic evaluation of deep learning models for vehicle classification and counting using numerical data
No full textThis study provides a comprehensive baseline evaluation of five deep learning models 1D Convolutional Neural Network (1DCNN), Long Short-Term Memory (LSTM), 2D Convolutional Neural Network (2DCNN), Recurrent Neural Network (RNN), and Autoencoder on two distinct dataset types: a synthetic dataset and a hybrid dataset. Structured numeric data exists in two separate datasets: The synthetic dataset spans from 10K to 100K points and the hybrid integration of synthetic data with real-world data also falls within this parameter range. Results from experiments show 1DCNN excels at numeric data processing due to its ability to deliver superior results with increased efficiency rates beyond competing models. The experimental results of this study validate the compatibility and efficient data handling abilities of 1DCNN across various large numeric
datasets, resulting in reduced computational complexity. This
study provides essential knowledge about deep learning model
capabilities in numeric data processing so future AI systems and data-driven decisions can develop further
Positive news culture: Flood framing in China
No full textFlood is an environmental hazard causing soil erosion, leading to not only damage of vegetation but also to water
contamination and spread of diseases. In China, the positive reporting culture shapes flood coverage by emphasising government responses and crisis management, reinforcing public trust and social stability. The study examined the flood framing by a national newspaper (People’s Daily) and a regional newspaper (Zhengzhou Daily) published in China. A total of 194 news articles on flood published between January 1 and December 31, 2023 were analysed. People’s Daily attributed more salience to flood news (179 articles) than Zhengzhou Daily (15 articles). The analysis showed that both newspapers predominantly utilised episodic framing (People’s Daily, 99.4 %; Zhengzhou Daily, 100 %). Through the positive valence of articles (70.9 %), People’s Daily set the agenda for the readers to showcase President Xi Jinping’s direct involvement in flood mitigation and highlight his compassion to victims. In contrast, Zhengzhou Daily had only 26.7 % of articles with positive valence. The larger percentage of articles with negative valence (53.3 %) alerts readers to the destruction and challenges caused by floods. The responsibility frame dominated in both newspapers. The second frequent frame dimension in People’s Daily is economic consequences of flood and whereas Zhengzhou Daily uses human interest elements to engage with readers. People’s Daily as the official voice box of the central government exudes propaganda even in crisis reporting
STRENGTH PREDICTION OF RSM IN BOLTED CONNECTION OF ALAN BATU WOOD
No full textThis study investigates the structural performance of bolted connections in Alan Batu wood to support the development of a local wood database for strengthening wall-diaphragm connections of unreinforced masonry (URM) buildings in Malaysia. Alan Batu wood was selected due to its common use as roof rafters and floor joists in URM buildings. Ten groups, each with ten replicates, were tested to assess bolted connection strength. Brittle failure was observed, confirming row shear failure. Since the material properties of Alan Batu are similar to Meraka wood, this study combines data from both to recommend design parameters (i.e., dry density and calibration factor) for optimising bolted connection strength predictions using the Row Shear Model (RSM). Analysis of the combined data showed that the Malaysian timber code (MS544-5) significantly underestimates bolted connection strength, potentially leading to over-sizing steel bolts and using more fasteners than necessary. The MS544-5 predictions had an average effectiveness of 45% compared to the 5th percentile experimental values, while the RSM showed a much higher average effectiveness of 80%. The design parameters proposed in this study optimise RSM strength predictions for Meraka and Alan Batu woods, offering cost-effective solutions for retrofitting URM wall-diaphragm connections
Bamboo-Based Soil Carbon Sequestration: A Sustainable Pathway for Soil Health and Climate Mitigation
No full textSoil carbon sequestration is a critical strategy for enhancing soil health and reducing the impact of climate change. The potential of bamboo to facilitate soil carbon storage and ecosystem restoration has garnered significant attention due to its high renewable potential and rapid
growth (Kotangale et al., 2025). Due to its extraordinary capacity to enhance below-ground carbon sequestration through substantial root biomass, soil organic matter accumulation, and microbial interactions, bamboo is a viable alternative for sustainable soil management (Nath et
al., 2015; Shoudho et al., 2024). As bamboo plantations have also been recognised for their ability to restore degraded land, they are an exceptional alternative to sustainable land-use planning
In vitro micropropagation and gas chromatography-mass spectrometry profiling of callus culture in Pulicaria jaubertii for conservation and metabolite production
No full textPulicaria jaubertii is an aromatic and medicinal plant endemic to Yemen, currently facing habitat decline. This study aimed to evaluate its in vitro response in full-strength Murashige and Skoog (medium supplemented with different types and concentrations of plant growth regulators. Among the tested plant parts, only seed explants successfully initiated callus formation. Calli were subsequently subcultured in media containing 0.1 mg/L 1-naphthaleneacetic acid (NAA) with kinetin (Kin) at 0, 0.25, 0.5, or 1 mg/L. Additional experiments tested media with 0.1 mg/L 6-benzylaminopurine and indole-3-acetic acid (IAA) (0–1 mg/L), as well as 0.1 mg/L Kin with 2,4-dichlorophenoxyacetic acid (2,4-D) (0–1 mg/L). Growth parameters related to callus induction, root, shoot, and leaf production were assessed. Findings revealed that Kin had no significant effect on most growth parameters except callus colour (P = 0.012), with the best growth at 0.25 mg/L. Similarly, IAA significantly influenced callus induction (P = 0.009), with optimal results at 1.0 mg/L. In contrast, 2,4-D had no significant effect, but its highest concentration (1.0 mg/L) supported optimal growth. Gas chromatography-mass spectrometry (GC-MS) analysis identified 46 compounds in the ethanolic callus extract compared to 25 in the mother plant, which indicates a richer phytochemical profile in the callus. The 2-Ethoxyethylamine (85.60%) and Stigmasterol (58.79%) were most abundant in ethanolic and n-hexane extracts. In conclusion, P. jaubertii seeds are the most responsive explants for micropropagation, forming callus as an initial step. Interestingly, GC-MS profiling identified bioactive compounds with medicinal properties. Further studies should refine auxin and cytokinin ratios to enhance propagation efficiency