http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Korean Text to Gloss: Self-Supervised Learning approach
Thanh-Vu Dang(Thanh-Vu Dang),김진영(JinYoung Kim),유광현(Gwang-Hyun Yu),김지용(Ji Yong Kim),박영환(Young Hwan Park),이칠우(ChilWoo Lee) 한국스마트미디어학회 2023 스마트미디어저널 Vol.12 No.1
Natural Language Processing (NLP) has grown tremendously in recent years. Typically, bilingual, and multilingual translation models have been deployed widely in machine translation and gained vast attention from the research community. On the contrary, few studies have focused on translating between spoken and sign languages, especially non-English languages. Prior works on Sign Language Translation (SLT) have shown that a mid-level sign gloss representation enhances translation performance. Therefore, this study presents a new large-scale Korean sign language dataset, the Museum-Commentary Korean Sign Gloss (MCKSG) dataset, including 3828 pairs of Korean sentences and their corresponding sign glosses used in Museum-Commentary contexts. In addition, we propose a translation framework based on self-supervised learning, where the pretext task is a text-to-text from a Korean sentence to its back-translation versions, then the pre-trained network will be fine-tuned on the MCKSG dataset. Using self-supervised learning help to overcome the drawback of a shortage of sign language data. Through experimental results, our proposed model outperforms a baseline BERT model by 6.22%.
Removing Out - Of - Distribution Samples on Classification Task
Thanh-Vu Dang,Hoang-Trong Vo,유광현(Gwang-Hyun Yu),이주환(Ju-Hwan Lee),Huy-Toan Nguyen,김진영(Jin-Young Kim) 한국스마트미디어학회 2020 스마트미디어저널 Vol.9 No.3
Out - of - distribution (OOD) samples are frequently encountered when deploying a classification model in plenty of real-world machine learning-based applications. Those samples are normally sampling far away from the training distribution, but many classifiers still assign them high reliability to belong to one of the training categories. In this study, we address the problem of removing OOD examples by estimating marginal density estimation using variational autoencoder (VAE). We also investigate other proper methods, such as temperature scaling, Gaussian discrimination analysis, and label smoothing. We use Chonnam National University (CNU) weeds dataset as the in - distribution dataset and CIFAR-10, CalTeach as the OOD datasets. Quantitative results show that the proposed framework can reject the OOD test samples with a suitable threshold.
Vu Dan Thanh Le,Anh Tuan Nguyen,Lac Hong Nguyen,Ngoc Thanh Dang,Ngoc Doan Tran,한재흥 한국항공우주학회 2021 International Journal of Aeronautical and Space Sc Vol.22 No.5
This paper explores the effectiveness of spin motion in mitigating the flight dispersion of a two-stage solid-propellant rocket model due to thrust misalignment. The aerodynamic coefficients of the rocket model are obtained by the use of a panel method and semi-empirical equations. A simulation program is developed to solve the equations of motion while considering the variations of the inertial parameters. Monte Carlo simulation techniques are applied to provide statistical data that are used to analyze the relationship between the spin motion and flight dispersion. The spin motion is generated by canting the fins to generate the axial aerodynamic moment. The results show that thrust misalignment at the first stage of the rocket has a great impact on the dispersion of rocket flight. By canting the first-stage fins at a relatively large angle to create the spin motion right after launch, the dispersion area of the payload-release location can be minimized considerably. However, thrust misalignment as well as the fin cant angle at the second stage appear to have insignificant effects on the rocket flight trajectory. On the other hand, canting the fins of the second stage at a large angle may lead to an increase in the spin rate, which may be harmful to the rocket operation. The paper also shows the variation of the dispersion characteristics of rocket flight when the fin size is modified.
Tran Ha Thi Thanh,Dang Anh Kieu,Ly Duc Viet,Vu Hao Thi,Hoang Tuan Van,Nguyen Chinh Thi,Chu Nhu Thi,Nguyen Vinh The,Nguyen Huyen Thi,Truong Anh Duc,Pham Ngoc Thi,Dang Hoang Vu 아세아·태평양축산학회 2020 Animal Bioscience Vol.33 No.10
Objective: The rapid and reliable detection of the African swine fever virus (ASFV) plays an important role in emergency control and preventive measures of ASF. Some methods have been recommended by FAO/OIE to detect ASFV in clinical samples, including real-time polymerase chain reaction (PCR). However, mismatches in primer and probe binding regions may cause a false-negative result. Here, a slight modification in probe sequence has been conducted to improve the qualification of real-time PCR based on World Organization for Animal Health (OIE) protocol for accurate detection of ASFV in field samples in Vietnam. Methods: Seven positive confirmed samples (four samples have no mismatch, and three samples contained one mutation in probe binding sites) were used to establish novel real-time PCR with slightly modified probe (Y = C or T) in comparison with original probe recommended by OIE. Results: Both real-time PCRs using the OIE-recommended probe and novel modified probe can detect ASFV in clinical samples without mismatch in probe binding site. A high correlation of cycle quantification (Cq) values was observed in which Cq values obtained from both probes arranged from 22 to 25, suggesting that modified probe sequence does not impede the qualification of real-time PCR to detect ASFV in clinical samples. However, the samples with one mutation in probe binding sites were ASFV negative with OIE recommended probe but positive with our modified probe (Cq value ranked between 33.12-35.78). Conclusion: We demonstrated for the first time that a mismatch in probe binding regions caused a false negative result by OIE recommended real-time PCR, and a slightly modified probe is required to enhance the sensitivity and obtain an ASF accurate diagnosis in field samples in Vietnam.
Thanh, Tran Dang,Linh, Dinh Chi,Yen, Pham Duc Huyen,Bau, Le Viet,Ky, Vu Hong,Wang, Zhihao,Piao, Hong-Guang,An, Nguyen Manh,Yu, Seong-Cho Elsevier 2018 PHYSICA B-CONDENSED MATTER - Vol.532 No.-
<P><B>Abstract</B></P> <P>In this work, we present a detailed study on the magnetic properties and the magnetocaloric effect (MCE) of La<SUB>1−x</SUB>K<SUB>x</SUB>MnO<SUB>3</SUB> compounds with <I>x</I>=0.05–0.2. Our results pointed out that the Curie temperature (<I>T</I> <SUB>C</SUB>) could be controlled easily from 213 to 306K by increasing K-doping concentration (<I>x</I>) from 0.05 to 0.2. In the paramagnetic region, the inverse of the susceptibility can be analyzed by using the Curie-Weiss law, <I>χ</I>(<I>T</I>)=<I>C</I>/(<I>T</I>−<I>θ</I>). The results have proved an existence of ferromagnetic clusters at temperatures above <I>T</I> <SUB>C</SUB>. Based on Banerjee's criteria, we also pointed out that the samples are the second-order phase transition materials. Their magnetic entropy change was calculated by using the Maxwell relation and a phenomenological model. Interestingly, the samples with <I>x</I>=0.1–0.2 exhibit a large MCE in a range of 282–306K, which are suitable for room-temperature magnetic refrigeration applications. The composites obtained from single phase samples (<I>x</I>=0.1–0.2) exhibit the high relative cooling power values in a wide temperature range. From the viewpoint of the refrigerant capacity, the composites formed out of La<SUB>1−x</SUB>K<SUB>x</SUB>MnO<SUB>3</SUB> will become more useful for magnetic refrigeration applications around room-temperature.</P>
Thanh-Tung Duong,Ta Quoc Tuan,Dang Viet Anh Dung,Nguyen Van Quy,Dinh-Lam Vu,Man Hoai Nam,Nguyen Duc Chien,윤순길,Anh-Tuan Le 한국물리학회 2014 Current Applied Physics Vol.14 No.12
Polyaniline nanowires (PANI NWs) were deposited onto fluorine-doped tin oxide (FTO) glass substrate using the cyclic voltammetric method with aniline monomer precursor in HCl aqueous solution. The secondary oxidation peak plays an important role in polymerization of aniline monomer and the optimization of catalytic activity of PANI-based counter electrodes was achieved by controlling the number of cycles. The photovoltaic performance of the dye-sensitized solar cells (DSSCs) with PANI NWs counter electrodes (CEs) was optimized at 4th cycles, and then following parameters were obtained: Jsc ¼ 17.2 mA cm2, Voc ¼ 0.71 V, FF ¼ 59.3%, and efficiency (h) ¼ 7.24%. While, Jsc ¼ 14.7 mA cm2, Voc ¼ 0.77 V, FF ¼ 70.6%, and efficiency (h) ¼ 7.98% in cells with Pt CEs. The PANI NWs were attractive as an alternative CEs for the low-cost DSSCs instead of Pt.
Split-and-Merge Fast Level Set Methods and their application to medical image segmentation
Vu Dang Tran,Young Jun Seo(서영준),Pham Thanh Trung,Jin Young Kim(김진영),Seoung You Na(나승유) 한국지능시스템학회 2009 한국지능시스템학회 학술발표 논문집 Vol.19 No.1
The boundary evolution based Level Set Method has been widely used in medical image segmentation. In this paper, we have developed a new algorithm that extends the Fast Level Set Method. The main contributions of this work are to formulate a new level set function, define relationships between layers and use an alternative elimination criterion in interface evolution. The experimental results validate the effectiveness and efficiency of our approach, compared to the classical Level Set Method and the original Fast Level Set Method.
Management of Antibiotic-Resistant Helicobacter pylori Infection: Perspectives from Vietnam
Vu Van Khien,Duong Minh Thang,Tran Manh Hai,Nguyen Quang Duat,Pham Hong Khanh,Dang Thuy Ha,Tran Thanh Binh,Ho Dang Quy Dung,Tran Thi Huyen Trang,Yoshio Yamaoka 거트앤리버 소화기연관학회협의회 2019 Gut and Liver Vol.13 No.5
Antibiotic resistance is the most important factor leading to the failure of eradication regimens. This review focuses on the prevalence of Helicobacter pylori primary and secondary resistance to clarithromycin, metronidazole, amoxicillin, levofloxacin, tetracycline, and multidrug in Vietnam. We searched the PubMed, EMBASE, Vietnamese National Knowledge Infrastructure, and Vietnamese Biomedical databases from January 2000 to December 2016. The search terms included the following: H. pylori infection, antibiotic (including clarithromycin, metronidazole, amoxicillin, levofloxacin, tetracycline, and multidrug) resistance in Vietnam. The data were summarized in an extraction table and analyzed manually. Finally, Excel 2007 software was used to create charts. Ten studies (three studies in English and seven in Vietnamese) were included in this review. A total of 308, 412, 523, 408, 399, and 268 H. pylori strains were included in this review to evaluate the prevalence of H. pylori primary resistance to amoxicillin, clarithromycin, metronidazole, levofloxacin, tetracycline, and multidrug resistance, respectively. Overall, the primary resistance rates of amoxicillin, clarithromycin, metronidazole, levofloxacin, tetracycline, and multidrug resistance were 15.0%, 34.1%, 69.4%, 27.9%, 17.9% and 48.8%, respectively. Secondary resistance rates of amoxicillin, clarithromycin, metronidazole, levofloxacin, tetracycline, and multidrug resistance were 9.5%, 74.9%, 61.5%, 45.7%, 23.5% and 62.3%, respectively. In Vietnam, primary and secondary resistance to H. pylori is increasing over time and affects the effectiveness of H. pylori eradication.
Novel reassortant H5N6 highly pathogenic influenza A viruses in Vietnamese quail outbreaks
Thanh, Hien Dang,Tran, Van Trung,Nguyen, Duc Tan,Hung, Vu-Khac,Kim, Wonyong Elsevier 2018 Comparative immunology, microbiology and infectiou Vol.56 No.-
<P><B>Abstract</B></P> <P>Avian influenza A H5N6 virus is a highly contagious infectious agent that affects domestic poultry and humans in South Asian countries. Vietnam may be an evolutionary hotspot for influenza viruses and therefore could serve as a source of pandemic strains. In 2015, two novel reassortant H5N6 influenza viruses designated as A/quail/Vietnam/CVVI01/2015 and A/quail/Vietnam/CVVI03/2015 were isolated from dead quails during avian influenza outbreaks in central Vietnam, and the whole genome sequences were analyzed. The genetic analysis indicated that hemagglutinin, neuraminidase, and polymerase basic protein 2 genes of the two H5N6 viruses are most closely related to an H5N2 virus (A/chicken/Zhejiang/727079/2014) and H10N6 virus (A/chicken/Jiangxi/12782/2014) from China and an H6N6 virus (A/duck/Yamagata/061004/2014) from Japan. The <I>HA</I> gene of the isolates belongs to clade 2.3.4.4, which caused human fatalities in China during 2014–2016. The five other internal genes showed high identity to an H5N2 virus (A/chicken/Heilongjiang/S7/2014) from China. A whole-genome phylogenetic analysis revealed that these two outbreak strains are novel H6N6-like <I>PB2</I> gene reassortants that are most closely related to influenza virus strain A/environment/Guangdong/ZS558/2015, which was detected in a live poultry market in China. This report describes the first detection of novel H5N6 reassortants in poultry during an outbreak as well as genetic characterization of these strains to better understand the antigenic evolution of influenza viruses.</P> <P><B>Highlights</B></P> <P> <UL> <LI> Two H5N6 influenza viruses were isolated from dead quails in central Vietnam in 2015. </LI> <LI> The whole virus genome was sequenced, and phylogenetic analysis was performed. </LI> <LI> A novel H6N6-like PB2 gene reassortant was identified from poultry. </LI> </UL> </P>