Tạp chí Khoa học và Công nghệ - Đại học Nguyễn Tất Thành

An artificial intelligence-based system for essay assessment at Nguyen Tat Thanh University

Pham Dinh Tai, Vuong Xuan Chi — Tue, 10 Feb 2026 07:33:18 +0700

In the context of modern higher education, manual essay grading has placed a tremendous burden on lecturers due to a huge volume of exams from large classes. This letter proposed an automated assessment system based on artificial intelligence (AI) intergrated with Google Drive for secure cloud storage and Streamlit for a friendly web interface to improve the efficiency and transparency of the learning process. To increase accuracy, pre-processing of words using TF-IDF algorithm was applied to extract core keywords from correct answers, followed by semantic comparison, cloud synchronization and intuitive interface. Results from real-world data testing proved that thanks to pre-processing, grading 30 papers was completed within 5 minutes, the deviation was reduced from (0,8 to 0,6) minutes and (300-450) minutes of grading time was saved compared to traditional grading method, thus achieving high reliability for large scale. Findings fron the present study confirmed that pre-processing method has increased accuracy by 25% by integrating Google Drive and Streamlit for comprehensive and scalable management, thereby solving Viet Nam's educational challenges.

Enhancing Image Captioning via Knowledge Graph Integration and R-CNN Network

Nguyen Kim Quoc, Dinh Xuan Thao, Dang Nhu Phu — Tue, 10 Feb 2026 09:58:43 +0700

In the era of digitalization, automatic image captioning plays a crucial role in image processing, while traditional methods are limited in understanding Semantic context. This study proposed a model which integrated R-CNN with knowledge graphs to enhance caption accuracy and completeness. The methodology involved constructing knowledge graphs from diverse sources like ImageNet and COCO, integrating into R-CNN via CNN feature extraction, Selective Search region proposals, softmax classification, and bounding box regression. Data collection used web crawlers and APIs, preprocessing removes noise and normalizes to 224 × 224. Training with 80:20 split, SGD optimizer, 0.001 learning rate, 50 epochs. Experimental results achieved 96 % accuracy and 0.75 average IoU on 2 000 test images, outperforming traditional R-CNN (85% accuracy; 0.6 IoU) due to graph semantics, reducing errors in complex images. Novel contribution of the present study included Semantic integration mechanism, improving flexibility for image management, healthcare, mobile apps. Future directions: optimizing complexity, multilingual support, combining with NLP like BERT. The research demonstrated expansion potential, recommended real-world application and dynamic graph updates. The computational complexity of the model increases slightly (~20%) compared to the traditional R-CNN, yet it still ensures real-time processing capability. When compared to Fast R-CNN and YOLO, the proposed model delivers significantly higher accuracy, despite a higher computational cost. This demonstrates a reasonable trade-off between performance and speed.

Numerical and experimental study on the flame flow field near the fuel nozzle tip

Tue, 10 Feb 2026 00:00:00 +0700

The generation of a recirculation zone in jet flow has recently been demonstrated to influence flame stability and soot formation. In this study, we conducted both experiments and numerical simulations to elucidate the mechanisms underlying vortex formation in laminar coflow diffusion flames. Numerical models were developed by incorporating various parameters, including fuel density, jet velocity, fuel temperature, and ambient air temperature. The flow patterns near the nozzle indicated how these parameters affect the development of the recirculation zone. Simulation results were validated against experimental data. Among the studied parameters, fuel density, jet velocity, and nozzle diameter emerged as the primary factors influencing recirculation flow, whereas fuel and ambient air temperatures exerted secondary effects. Detailed analysis of each parameter unveiled critical conditions at which the recirculation zone forms.

Automation Solution for Railway Crossing Safety Control: A Study on Automatic Barrier Control System

Pham Minh Tri, Ho A Lil, Nguyen Huu Luan, Bui Vu Minh — Tue, 10 Feb 2026 09:57:13 +0700

This study focuses on the design and implementation of an automated control system aimed at optimizing safety and mitigating accident risks at unprotected railway-road intersections. The model was constructed based on a three-station distributed embedded control architecture, utilizing microcontrollers and a LoRa wireless communication network. Station 1 employed geophone sensors for seismic data acquisition to determine the presence of an approaching train. Station 2 served as the coordinator, transmitting early warning commands. Station 3, situated at the crossing, was responsible for barrier mechanism control and integrates computer vision for monitoring and detecting intrusion into the danger zone. The novel contribution is the synthesis of physical sensors and an intelligent vision system. This model incorporated a bidirectional safety mechanism, enabling Station 3 to reverse-alert the train (via Station 2) upon detecting unsafe obstacles. Experimental results demonstrated the system’s reliable operation, with a successful signal transmission/reception rate between stations reaching (1.2–1.8) km. Furthermore, the system exhibited low power consumption, quantified at approximately 81 Wh/24 hours. When connected with a 6750 mAh backup battery and photovoltaic cells, the stations were capable of continuous and autonomous operation without reliance on the power grid.

Optimization of synthesis conditions of NiFe2O4@AC from Bidens pilosa using response surface methodology for crystal violet dye adsorption

Duyen Thi Cam Nguyen — Tue, 10 Feb 2026 09:56:53 +0700

Bidens pilosa is a harmful invasive species that competes fiercely with crops, yet thanks to its vigorous growth, it also provides an abundant natural biomass source. This study focused on the synthesis of a magnetic nanocomposite, NiFe₂O₄@AC derived from B. pilosa biomass for the removal of Crystal Violet dye. Using the Response Surface Methodology (RSM), the synthesis conditions were successfully optimized. The predictive model identified the optimal parameters for maximum adsorption capacity as: 28.82 % NFO loading, 122.89 °C temperature, and 11.88 hours reaction time. Under these conditions, the optimized NiFe₂O₄@AC exhibited excellent dye removal efficiency, with an experimental adsorption capacity of 96.87 mg/g, confirming its potential for wastewater treatment.

Comparison of growth and yield of Basil (Ocimum basilicum) cultivated in aquaponics (with Oreochromis sp.) and recirculating hydroponic systems in Ho Chi Minh City

Tran Thi Quy, Tran Thi Huong, Bui Le Trong Nhan, Ngo Thi Lam Giang — Tue, 10 Feb 2026 09:55:21 +0700

Aquaponics is a promising approach to sustainable agriculture by integrating vegetable cultivation and aquaculture in a circular system that optimizes resource use. This experiment aimed to compare the growth and yield of basil (Ocimum basilicum) cultivated in two aquaponics systems combined with Oreochromis sp.: NFT (Nutrient Film Technique) and MB (Media Based), with a conventional recirculating hydroponics system. Results showed that water quality parameters in aquaponics systems remained within suitable ranges for both plants and fish. Basil yields in NFT (7.0 kg∙m^-2) were comparable to hydroponics (7.3 kg∙m^-2) and higher than MB (5.13 kg∙m^-2). Oreochromis sp. raised in aquaponics systems achieved yields of (34.39-34.60) kg∙m^-3, outperforming the control system with fish only (25.6 kg∙m^-3). These findings demonstrate that the NFT aquaponics system can effectively replace recirculating hydroponics for basil production while enhancing fish growth performance, representing a sustainable solution for agricultural development.

Preparation, Quality Evaluation, and Determination of Antioxidant and in vitro Anti-inflammatory Activities of 5-herb Extract

Nguyen Ngoc Sao Mai, Le Thi My Linh, Le Minh Tri, Dang Minh Khoi — Tue, 10 Feb 2026 09:54:57 +0700

“Ngũ vị” is a herbal remedy from Truong Giang Traditional Medicine Clinic specialized in treating various coughs. The abundant saponin content in the remedy has anti-inflammatory, antioxidant, mucolytic, and respiratory membrane-protective effects. To ensure convenience, Ngũ Vị concentrated extract was prepared and evaluated according to the requirements of the Vietnamese Pharmacopoeia V (VP V), determining total ash content, performing qualitative and quantitative analyses, and assessing some in vitro biological activities such as antioxidant activity by DPPH method and anti-inflammatory activity. The results showed that Ngũ Vị concentrated extract met the criteria of VP V, with a total ash content of 3.41 %, steroidal saponin content of (32.02 ± 0.81) μg/mL, antioxidant capacity IC₅₀ of 32.04 mg/mL (compared to ascorbic acid 7.44 μg/mL), and anti-inflammatory capacity IC₅₀ of 6.92 mg/mL (compared to diclofenac with 489.25 μg/mL). Ngũ vị concentrated extract can be applied as a supportive preparation in treating persistent respiratory diseases due to "heat in the lung" or "qi-yin deficiency" factors in traditional medicine, from which it can be developed into modern dosage forms such as medicinal syrups, capsules, and instant teas.

Processing and utilization powder from keo mango peel and seed kernels in biscuit products

Diep Dang Hung, Duong Thi Phuong Lien — Tue, 10 Feb 2026 09:54:34 +0700

Mango is a widely consumed fruit and is extensively cultivated in Mekong Delta. Mango peel and kernel contain significant nutritional value and bioactive compounds with strong antioxidant properties; however, they are often discarded during processing. This study utilized Keo mango peel and kernel by-products to produce value-added semi-finished materials for incorporation into food products. The research investigated the effect of replacing wheat flour with a mixture of mango peel powder and kernel powder at various substitution levels of (0; 5; 10; and 15) %, and peel-to-kernel ratios (0/100; 25/75; 50/50; 75/25; and 100/0) to determine the appropriate mixture that maintains desirable structural, quality, and bioactive properties. Results indicated that replacing 10 % of wheat flour with a 50/50 blend of Keo mango peel and kernel powders yielded cookies with favorable sensory characteristics. The cookies had a moisture content of (1.38 ± 0.08) %, spread ratio of (35.40 ± 1.67) %, hardness of (6947.1 ± 176.5) g, total polyphenol content of (148.80 ± 1.58) mg GAE/g, and DPPH radical scavenging activity of (78.68 ± 2.65) %. These findings demonstrated the potential application of mango peel and kernel powders in food processing, contributing to the enhancement of by-product value and the reduction of environmental pollution.

Study on the effects of organic fertilizer solution and shading level on the growth of Azolla microphylla

Bui Le Trong Nhan, Tran Thi Quy, Tran Thi Huong — Tue, 10 Feb 2026 09:54:08 +0700

In the trend of sustainable agriculture, the use of agricultural by-products to replace chemical fertilizers is increasingly focused. Azolla sp. is a green manure source which is rich of protein, minerals and able to fix symbiotic nitrogen that contributes to cost reduction and soil’s quality improvement. This study evaluated the effect of organic solutions derived from fish’s by-products, cow dung and Azolla, combining effect of different shaded ranges on the growth of Azolla microphylla. The experiment were arranged in a completely random block design (RCBD) using styrofoam boxes (32 × 45 × 20) cm with 10 cm depth of water, adding with 5 g of Azolla microphylla each box. Results showed that the 50-fold diluted organic solution derived from fish’s by-products promoted the best growth of Azolla microphylla. 40 % of shade was confirmed suitable for Azolla microphylla growth after 7 days, leaves’ horizontal length reached 12.3 mm, fresh biomass per box reached 68.55 g, dry biomass per box reached 5.81 g, and dry matter reached 8.48 %. The research proved the potential applying of fish’s by-products in production of other crops, thereby promoting the circular economy in agriculture.

The Significance of Relief and Ornament in the Architecture of Huế's Royal Complex

Ngo Ngan Ha — Tue, 10 Feb 2026 09:53:46 +0700

Visual arts in general, relief and ornament in particular, within the architecture of Hue’s Royal Complex during the Nguyen Dynasty is one of the essential components contributing to the comprehensive values of the Complex of Hue Monuments. In addition to their placement in various locations on architectural structures, relief and ornament are expressed through a wide range of thematic systems, showcasing rich diversity and high aesthetic value. This study identified the characteristics of relief and ornament through elements such as theme, color, composition, technique, and material, as manifested in both the interior and exterior of selected architectural works within Hue’s Royal Complex. The research applied a combination of methods, including synthesis, secondary source analysis, art-historical approaches, iconographic analysis, field surveys and measurements, as well as replication of ornamental motifs on site. The findings revealed that the system of relief and ornament in Hue’s Royal Complex was not merely an assemblage of decorative elements but a deeply encoded visual language embodying the worldview and ideological framework of the Nguyen Dynasty. These results not only confirmed the aesthetic values but also established their significance and role as important historical sources within the cultural continuum of Viet Nam. Consequently, the study underscored the necessity of conservation and promotion efforts to preserve this heritage as a constituent part of the national identity for future generations.

Applying the Values of Buffer Spaces in Traditional Architecture to Residential Villas (A Case Study of a Villa in Buon Ma Thuot)

Dinh Thanh Ngan — Tue, 10 Feb 2026 09:53:15 +0700

In the context of contemporary urban housing development, modern design often emphasizes comfort and convenience but gradually loses traditional values such as adaptation to nature, family cohesion, and community bonding. This study aimed to identify the characteristics of buffer spaces in traditional Vietnamese architecture and assess their applicability in the design of urban villas, with a case study of a villa in Buon Ma Thuot City. The research methodology combined literature review and analysis, field surveys, and comparative evaluation of the case study against traditional buffer space models from Northern, Central, and Southern regions of Viet Nam. The findings highlighted the core values of buffer spaces in organizing Vietnamese living environments, including regulating the microclimate, enhancing spatial connectivity, and preserving cultural values. Then, the study evaluated the extent to which these elements have been incorporated into the selected villa, thereby proposing directions for developing modern villa architecture that are sustainable and rooted in local traditions.