• Journal of The Korean Association For Science Education
• /
• v.35 no.4
• /
• pp.649-663
• /
• 2015

This study investigated the current state of writing in research papers of science education with focus on the translationese and basic Korean grammar, and found a way of improving the Korean language. The science education research have characteristics of both social science and natural science, and of having more quantitative than qualitative research, which could influence the writing of the research paper. The translationese means the conventional expression originated from foreign language other than Korean. The basic Korean grammar includes 'agreement,' 'spelling, word spacing, punctuation mark,' 'causative suffix,' 'use of English or loanword,' and the translationese is divided in 'English,' 'Japanese,' and 'English and Japanese.' The sentences in nine research papers in the 'Journal of the Korean Association for Science Education' were analyzed, and the problematic sentences were discussed and provided with alternatives. The cases with high frequency include '-jeok,' 'use of English,' 'expression of the plural,' 'passive voice of the verb with -hada,' '-go inneun,' '-eul tonghayeo,' '-e daehayeo,' 'gajida,' 'genitive case marker -eui,' 'passive voice with subject of thing,' and 'causative suffix, -sikida.' Based on the results, the characteristics of writing of science education research papers were described as 'writing of quantitative research,' 'objective writing of academic research,' and 'writing of research of foreign origin.' In order to improve the writing of research paper of science education, the science education researcher should pay attention to basic Korean grammar and the translationese, and be familiar with the concrete examples of problematic cases. The results of this study could be used in the education of writing and grammar of Korean language.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.01a
• /
• pp.61-63
• /
• 2019

본 논문에서는 웹툰 및 웹툰 플랫폼이 효율적으로 국외 시장에 진출하거나 또는 작품 원문의 언어를 사용하지 않는 고객에게 효과적으로 서비스 할 수 있도록 크라우드 소싱을 이용한 번역 및 투고 시스템과 이를 적용한 서비스 매체를 제안한다. 제안하는 시스템은 콘텐츠 번역 자원을 크라우드 소싱을 통해 효율적으로 확보한다. 이를 통해 콘텐츠를 세계화하는데 투입되는 비용을 최소화한다. 또한 디지털 콘텐츠가 디지털 매체를 통해 제공되므로 소프트웨어를 응용 할 수 있는 환경 특성을 이용하여 효과적으로 콘텐츠 내용을 관리하고 검증 할 수 있다. 본 논문에서 제안하는 크라우드 소싱을 활용한 번역 및 투고 시스템을 적용한 디지털 콘텐츠 및 서비스 매체는 기존 디지털 콘텐츠 및 서비스 매체에 비해 콘텐츠 세계화에 필요로 하는 비용효율 면에서 우수함을 보인다.

• Journal of the Korea Convergence Society
• /
• v.12 no.11
• /
• pp.69-80
• /
• 2021

Neural machine translation using deep neural network has emerged as a mainstream research, and an abundance of investment and studies on model structure and parallel language pair have been actively undertaken for the best performance. However, most recent neural machine translation studies pass along decoding strategy to future work, and have insufficient a variety of experiments and specific analysis on it for generating language to maximize quality in the decoding process. In machine translation, decoding strategies optimize navigation paths in the process of generating translation sentences and performance improvement is possible without model modifications or data expansion. This paper compares and analyzes the significant effects of the decoding strategy from classical greedy decoding to the latest Dynamic Beam Allocation (DBA) in neural machine translation using a sequence to sequence model.

• Journal of the Korea Convergence Society
• /
• v.12 no.5
• /
• pp.23-29
• /
• 2021

Recent deep learning-based natural language processing studies are conducting research to improve performance by training large amounts of data from various sources together. However, there is a possibility that the methodology of learning by combining data from various sources into one may prevent performance improvement. In the case of machine translation, data deviation occurs due to differences in translation(liberal, literal), style(colloquial, written, formal, etc.), domains, etc. Combining these corpora into one for learning can adversely affect performance. In this paper, we propose a new Corpus Weight Balance(CWB) method that considers the balance between parallel corpora in machine translation. As a result of the experiment, the model trained with balanced corpus showed better performance than the existing model. In addition, we propose an additional corpus construction process that enables coexistence with the human translation market, which can build high-quality parallel corpus even with a monolingual corpus.

• Journal of Plant Biotechnology
• /
• v.47 no.4
• /
• pp.316-323
• /
• 2020

This study aimed to evaluate the potential allergenicity and oral toxicity of the phosphinothricin acetyltransferase (PAT) protein expressed in Zoysia japonica, a herbicide-tolerant genetically modified (GM) zoysiagrass. In silico analysis of PAT showed no similarities with any known allergenic or toxic proteins, with <35% amino acid sequence homology with known allergens across a length of 80 amino acids and no continuous eight amino acid identity with known allergens. The PAT protein expressed in Z. japonica degraded very rapidly in the simulated gastric fluid in the presence of pepsin, and, no glycosylation of PAT was observed. The oral toxicity test revealed no mortality or toxic effect in mice following PAT administration at 4,000 mg/kg body weight. Our findings indicate that the PAT protein expressed in Zoysia japonica does not exhibit allergenic or toxic properties.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.3
• /
• pp.125-132
• /
• 2022

Sentence compression is a natural language processing task that generates concise sentences that preserves the important meaning of the original sentence. For grammatically appropriate sentence compression, early studies utilized human-defined linguistic rules. Furthermore, while the sequence-to-sequence models perform well on various natural language processing tasks, such as machine translation, there have been studies that utilize it for sentence compression. However, for the linguistic rule-based studies, all rules have to be defined by human, and for the sequence-to-sequence model based studies require a large amount of parallel data for model training. In order to address these challenges, Deleter, a sentence compression model that leverages a pre-trained language model BERT, is proposed. Because the Deleter utilizes perplexity based score computed over BERT to compress sentences, any linguistic rules and parallel dataset is not required for sentence compression. However, because Deleter compresses sentences only considering perplexity, it does not compress sentences by reflecting the linguistic information of the words in the sentences. Furthermore, since the dataset used for pre-learning BERT are far from compressed sentences, there is a problem that this can lad to incorrect sentence compression. In order to address these problems, this paper proposes a method to quantify the importance of linguistic information and reflect it in perplexity-based sentence scoring. Furthermore, by fine-tuning BERT with a corpus of news articles that often contain proper nouns and often omit the unnecessary modifiers, we allow BERT to measure the perplexity appropriate for sentence compression. The evaluations on the English and Korean dataset confirm that the sentence compression performance of sentence-scoring based models can be improved by utilizing the proposed method.

• Applied Biological Chemistry
• /
• v.39 no.6
• /
• pp.443-448
• /
• 1996

In order to investigate the function of xylA promoter(Pxyl) as regulatory region Pxyl-lacZ fusion gene was constructed by the insertion of xylA promoter to the multiple cloning site of upstream of lacZ gene in a multicopy numbered plasmid pMC1403 containing promoterless lac operon, which was designated pMCX191, and Pxyl-lacZ fragment from pMCX191 was inserted to low copy numbered plasmid pLG339, designated pLGX191. The expressions of ${\beta}-galactosidase$ in these recombinant plasmids containing Pxyl-lacZ fusion gene were induced strongly by the addition of xylose, repressed by the addition of 0.2% glucose in the presence of xylose. The catabolite repressions were derepressed by the addition of 1 mM cAMP as same as native xylA gene. The fragment of xylA promoter was partially deleted from the upstream of xylA promoter by exonuclease III to investigate the regulation site of xylA promoter and the degrees of deletion derivatives of xylA promoter were analyzed by the DNA base sequencing. By the investigations of the induction by xylose, repression by glucose and derepression by cAMP on xylose isomerase production, the regulation site of xylA promoter may be located in segment between -165 and -59 bp upstream from the initiation site of xylA translation.

• Journal of Life Science
• /
• v.33 no.3
• /
• pp.242-251
• /
• 2023

The purpose of this study was to investigate the effects of 8-week β-hydroxybutyrate (β-HB) administration with and without endurance exercise training on endurance exercise performance and skeletal muscle protein synthesis and degradation using a mouse model. Forty-eight male wild-type ICR mice (8 weeks old) were randomly divided into four groups: sedentary control (Sed+Con), (Sed+Con), sedentary β-HB (Sed+β-HB), exercise control (Exe+Con), and exercise β-HB (Exe+β-HB). β-HB was dissolved in PBS (150 mg/ml) and injected subcutaneously daily (250 mg/kg) for 8 weeks. Mice performed 5 days/week of a 20 min treadmill running exercise for 8 weeks. The running exercise was carried out at a speed of 10 m/min at a 10° incline for 5 min, and then the speed was increased by 1 m/min for every 1 min of the remaining 15 min. Following 8 weeks of treatments, visceral fat mass and skeletal muscle mass, blood parameters, and the markers for autophagy and protein synthesis were analyzed. The data were analyzed with one-way ANOVA (p<0.05) using the SPSS 21 program. Eight weeks of Exe+β-HB treatment significantly lowered blood lactate levels compared with the other three groups (Sed+Con, Sed+β-HB, and Exe+β-HB) Exe+β-HB) (p<0.05). Eight weeks of Exe+β-HB significantly increased maximal running time (time to exhaustion) compared with the Sed+Con and Exe+Con groups (p<0.05). Eight weeks of β-HB administration significantly decreased autophagy flux and autophagy-related proteins in the skeletal muscle of mice (p<0.05). Conversely, the combined treatment of β-HB and endurance exercise training increased protein synthesis (mTOR signaling and translation) (p<0.05). The 8-week β-HB treatment and endurance exercise training had synergistic effects on the increase in endurance performance, increase in protein synthesis, and decrease in protein degradation in the skeletal muscle of mice.