Ho Chi Minh City Open University Journal of Science
Not a member yet
16613 research outputs found
Sort by
Optimization of indole-3-acetic acid production by Bacillus subtilis strain IA3
Get PDFIndole-3-Acetic Acid (IAA) is a crucial plant growth-promoting hormone produced as a secondary metabolite by various microorganisms. The present study aims to isolate and screen IAA-producing endospore-forming bacteria from soil samples collected in Vietnam. A total of ten bacteria were isolated, and they were shown to have the ability to produce IAA, with strain IA3 exhibiting the highest production. Strain IA3 was identified as Bacillus subtilis based on the 16S rRNA sequencing and subsequently used to optimize conditions for IAA production. To maximize IAA yield, key factors influencing production were analyzed using Box-Behnken design. The optimized conditions led to a maximum IAA production of 47.67 ± 0.78 g/mL under the following conditions: 15.65 g/L molasses, 1.06 g/L (NH4)2SO4, and 0.52 g/L tryptophan. Therefore, Bacillus subtilis IA3 has potential applications for stimulating and enhancing plant growth or crop production
Association of SNP Rs34678647 with breast cancer risk in the Vietnamese population: An initial study
Get PDFBreast cancer remains a leading cause of mortality among women worldwide. In Vietnam, the rising incidence underscores the urgent need for early diagnostic markers. Genetic factors, particularly Single Nucleotide Polymorphisms (SNPs), play a critical role in breast cancer susceptibility. The SNP rs34678647, located downstream of the miR-221/222 cluster targeting Estrogen Receptor alpha (ERα), is hypothesized to influence breast cancer risk by modulating cancer-related pathways. This study investigates the association between rs34678647 and breast cancer risk in the Vietnamese population. A total of 234 DNA samples, comprising 131 breast cancer cases and 103 healthy controls, were genotyped using the Polymerase Chain Reaction-High Resolution Melting (PCR-HRM) technique, which is selected for its high sensitivity and specificity in SNP detection. Genotype frequencies were determined, and statistical analyses were conducted to evaluate associations. The PCR-HRM method successfully genotyped rs34678647 with high accuracy. The T allele was observed in 23% of cases and 19% of controls; however, no significant association with breast cancer risk was identified (OR = 1.19, 95% CI: 0.77-1.85, p = 0.43). Genotype distributions conformed to the Hardy-Weinberg equilibrium in both groups, supporting the representativeness of the sample. In conclusion, while the T allele of rs34678647 showed a nonsignificant trend toward increased breast cancer risk among Vietnamese women, the results were not statistically conclusive. These findings highlight the need for larger-scale studies to further explore the potential role of rs34678647 as a genetic risk factor and its applicability as an early diagnostic biomarker
Hybrid AI Model for User Emotion Recognition in F&B
Get PDFThis study proposes a hybrid approach combining facial emotion recognition and textual sentiment analysis to evaluate customer behavior in the food and beverage (F&B) industry. The model utilizes facial images to detect visual emotions and textual feedback to analyze sentiment, providing a comprehensive view of the customer experience. Experimental results show that the facial emotion recognition model achieved an accuracy of 69.4% on the FER2013 dataset and 91.1% on the Emodata dataset (a self-collected dataset). Meanwhile, the sentiment analysis model reached an accuracy of 91.2% on the fbr-vn dataset. These results demonstrate the effectiveness of the proposed approach in identifying emotions, highlighting its potential applications in helping businesses enhance service quality, improve customer experience, and optimize business strategies
Combining NLP and OCR for Multilingual Plagiarism Detection: An English–Vietnamese Case Study
No full textThis paper presents the development of a comprehensive multilingual plagiarism detection system that combines Natural Language Processing (NLP) and deep learning to process both textual and image-based inputs. By fine-tuning a Multilingual E5 embedding model specifically for English–Vietnamese semantic similarity tasks and integrating Optical Character Recognition (OCR) for extracting text from images, the system is able to detect both cross-language and cross-format plagiarism. Experimental results on semantic similarity datasets indicate that the proposed approach achieves competitive performance on semantic similarity datasets
Assessing the cooling effect of the blue corridors in Antakya pre-earthquake, Türkiye
No full textBlue corridors provide nature-based solutions in urban areas by reducing air temperatures in their surroundings. This study examines the impact of blue corridors on surface temperature and their relationship with land cover in pre-earthquake Antakya. The cooling effect of rivers, based on distance and density across different land cover types, was determined using NDVI, NDWI, and NDMI indices, and the relationship between these indices and LST was analyzed statistically. The results clearly show that the measured temperature difference of 15.421 °C in the corridors and their surroundings during the April–September period of 2020–2022 indicates microclimatic variations influenced by land cover characteristics. Although the corridors don't exhibit a distinct cooling effect in terms of water presence, the cooling impact of areas with dense vegetation remains limited. Therefore, increasing permeable surfaces and adopting water-sensitive planning and design approaches in and around the corridors are crucial for extending the cooling effect over a larger area
Attitude markers and self-mentions in research article discussions: A comparative analysis of applied linguistics and civil engineering disciplines
No full text
