Le Hong Phong High School for the Gifted

Growing concerns about environmental sustainability and ethical business practices have significantly influenced consumer behavior in recent years. This study investigates the impact of ethical consumerism on consumers’ purchase intention toward natural products in Vietnam, with brand trust serving as a mediating factor. Drawing on the Theory of Planned Behavior and Signaling Theory, the research examines three key dimensions of ethical consumerism, including sustainability perception, supply chain transparency, and ethical production perception. 311 data of young consumers were collected from consumers in Vietnam and analyzed using structural equation modelling with SmartPLS 3 and SPSS 27. The results indicate that sustainability perception, supply chain transparency, and ethical production perception all have significant positive effects on brand trust, with ethical production perception showing the strongest influence. Brand trust is also found to significantly influence purchase intention toward natural products and mediates the relationships between ethical consumerism factors and purchase intention. In addition, the study identifies significant differences in purchase intention across generational groups, individualism–collectivism orientation, and levels of exposure to foreign environments. These findings highlight the important role of ethical perceptions and trust in shaping consumer behavior in emerging markets. The study contributes to the literature on ethical consumerism by providing empirical evidence from Vietnam and offers practical insights for businesses seeking to strengthen consumer trust and promote sustainable consumption.

The study aims to evaluate the impact of supplementing with Combi VPlantex micronutrients on the growth, yield and silymarin content in milk thistle seeds. The experiment consisted of four treatments, arranged in a completely randomized design. The study monitored growth indicators, yield components, yield and silymarin content in milk thistle seeds. The results have shown that the Combi VPlantex micronutrient had a positive impact on the growth, development and yield of the milk thistle grown in Phu Tho. Combi Vplantex fertilizer sprayed at a concentration of 15% yielded outstanding results with an individual seed yield of 12.8 g/plant, an actual seed yield of 393.8 kg/ha and a silymarin content of 2.39%. However, spraying Combi VPlantex at a concentration of 12.5% also yielded results equivalent to those at a concentration of 15% with 95% reliability, indicating that this could be the optimal concentration for practical production of milk thistle to ensure high economic efficiency while still achieving good yield and quality.

This paper proposes a Deep Q-Network-based reinforcement learning framework to optimize the selection of technical trading strategies for Bitcoin. Instead of directly predicting prices, the agent chooses among five predefined strategies—RSI, SMA Crossover, Bollinger Bands, Momentum-20d, and VWAP Reversion—allowing actions to remain interpretable while adapting dynamically to market conditions. Market states are constructed from technical indicators, normalized, and reduced in dimensionality using Principal Component Analysis to ensure robustness and efficiency. Empirical tests on Bitcoin data from 2022 to mid-2025 show remarkable performance: starting from $1,000,000, the agent achieved more than 120-fold growth in Net Asset Value, significantly outperforming Buy-and-Hold and single-strategy benchmarks. The model flexibly emphasized momentum during uptrends, limited mean-reversion during breakouts, and managed drawdowns when temporarily misaligned. Out-of-sample results further confirmed generalization. Overall, the framework demonstrates a practical, explainable, and scalable approach to algorithmic trading in digital asset markets.

This study presents a comparative analysis of Long Short-Term Memory (LSTM) and Transformer models for stock price prediction across datasets of varying lengths. Using the QQQ stock dataset over two distinct periods, 7 months and 13 years, the research evaluates the models based on accuracy, training and inference time, and GPU memory consumption. Results indicate that LSTM slightly outperforms Transformer on shorter datasets due to its simpler architecture and lower computational demands. However, the Transformer exhibits significant advantages on longer datasets, offering faster training, up to three times quicker, and better scalability through its parallel processing and self-attention mechanisms. The findings highlight the Transformer’s superior performance on large-scale time series data, where it maintains accuracy while avoiding issues like the vanishing gradient problem commonly seen in LSTM. This underscores the Transformer’s suitability for long-term financial forecasting tasks, particularly when sufficient computational resources are available.

Memristors have been employed in the design of integrated memory, logic gates, analog circuits, and neuromorphic systems. Their non-volatile nature, high integration density, and compatibility with CMOS technology have been widely documented in numerous studies. Existing research has compared the operational characteristics of various memristor models, including linear, nonlinear, current-controlled threshold adaptive models, and voltage-controlled adaptive models. However, many of these traditional models are either too domain-specific or physically complex, which limits their practical simulation efficiency. In response to the need for a simple, general-purpose, and fast-simulating model in EDA flows, this study proposes an BV-VTEAM which is advanced version of the V-VTEAM model for bipolar memristors. The new model maintains a balance between simplicity and accuracy. In the BV-VTEAM model, the internal state variable represents the memristor's conductivity level. This dynamic variable evolves over time based on the applied stimulus. The state update is performed using a manual integration method, where the time step is automatically selected based on the nearest absolute timestamps in both the growth and decay phases. The bounded function is applied to ensure that internal state variable remains within physical limits. Additionally, the output current is regulated by a maximum threshold using a soft-bounded approach to avoid Zero Jacobian issues, thus improving numerical stability. The operation conditions are analyzed in various dynamic parameters in terms of supply voltage, operating frequency, maximum current, time constant, and nonlinearity exponent This BV-VTEAM model is further utilized to design digital logic circuits, including AND, OR, and flip-flops, laying the groundwork for constructing more complex digital systems based on memristive components.

Glioblastoma is a brain tumour with limited therapeutic options, prompting the exploration of novel treatments. This study evaluates the effects of α-mangostin, a xanthone derived from the pericarp of mangostin (Garcinia mangostana L), on C6 glioma cells. α -mangostin was successfully purified with a purity of ≥ 97%, and its chemical structure was confirmed. This study evaluated the antioxidant and anticancer activities of α-mangostin isolated from G. mangostana. The compound showed moderate antioxidant capacity (DPPH• IC50 = 67.55 ± 0.91 μg/mL) and potent cytotoxicity against C6 glioma cells (IC50 = 4.67 μg/mL). Mitochondrial membrane potential dropped to 34.27 ± 1.24% and 25.77 ± 0.98% at 5 and 8 μg/mL, respectively. DNA fragmentation increased from 2.17 ± 1.543% (control) to 17.95 ± 6.277% at 5 μg/mL. These findings suggest that α-mangostin induces glioma cell death via mitochondrial dysfunction and genotoxic stress, supporting its potential as a therapeutic lead for glioblastoma.

This study investigates the determinants and outcomes of business intelligence (BI) adoption within Vietnam's rapidly evolving retail sector, addressing critical gaps in understanding how technological capabilities drive strategic performance enhancement in emerging economies. Employing a mixed-method approach combining structural equation modelling (SEM) with fuzzy-set qualitative comparative analysis (fsQCA), the research examines data from 312 retail enterprises across Vietnam's major urban centres. The theoretical framework synthesises the technology-organisation-environment (TOE) framework with dynamic capabilities theory to explicate the complex pathways through which BI adoption influences operational efficiency, customer relationship management, and competitive advantage. The findings reveal that technological readiness, organisational culture, and environmental complexity collectively explain 68% of the variance in BI adoption intensity, whilst BI capabilities demonstrate significant positive effects on strategic performance outcomes (β = 0.742, p < 0.001). The fsQCA results identify four distinct configurational pathways to high performance, suggesting that successful BI implementation requires synergistic combinations of technological infrastructure, managerial support, and environmental alignment. This research contributes to the literature by advancing a comprehensive theoretical model that integrates institutional theory with technological diffusion perspectives, whilst providing practical insights for retail executives navigating digital transformation in emerging markets. The study's implications extend beyond Vietnam's retail context, offering valuable frameworks for understanding BI adoption patterns across developing economies experiencing rapid technological modernisation.

Hemidactylus frenatus Duméril & Bibron (Gekkonidae) is a fairly common species and has a wide distribution from Asia, Americas, Africa, Australia, and a number of tropical islands worldwide. Although this species is distributed from the North to the South in Vietnam, studies on morphological characters and genetic diversity are limited. In this study, we used two mtDNA sequences, Cytb (Cytochrome b) and ND2 (NADH dehydrogenase subunit 2), to evaluate the genetic relationships of five H. frenatus specimens in Nghe An province. The genetic distances between the studied samples were relatively low, 0.3% and 0.1% for Cytb and ND2 sequences, respectively. Phylogenetic analyses based on Cytb and ND2 sequences showed that all five samples were grouped together with H. frenatus on GenBank with significant support (100%). The genetic divergence between the H. frenatus specimens in Nghe An and the H. frenatus specimens on GenBank were also small, ranging from 0.0% to 0.6%, while high genetic distance between the five H. frenatus specimens and othe species in the genus Hemidactylus was observed, ranging from 29% to 46%. In addition, the morphological characters of H. frenatus species in Nghe An are also described to provide additional biological data for studies on genetic relationships and the invasion process of the species.