• Korean Journal of Human Ecology
• /
• v.1 no.1
• /
• pp.44-65
• /
• 1998

The purpose of this study are to measure the overall level of consumer consciousness and behavior on environmental problems of high school girl students and to analyze the most influential factors related to the socio-demography and the education on environment. This study also aims at inspiring the students with sound view of environments and furnishing basic data for the development of consumer education program. The researcher used questionnaires on consumer consciousness scale and consumer behavior scale for the purpose of this study. 640 students were surveyed randomly in Kwangju and Chonnam, and 608 questionnaires were selected as final analysis data. Major Findings were as follows: 1. The average score of consumer consciousness and behavior on environmental problems of the high school girls were 86.0 on scale of 110 (72.7 on scale of 100) and 63.3 on scale of 115(43.8 on scale of 100). 2. The consumer consciousness and behavior on environmental problems of the high school girls showed partial differences according to the socio-demography variable. 3. The consumer consciousness and behavior on environmental problems of the high school girls showed partial differences according to the education to environment variables. 4. The consumer behavior on environmental problems had a positive relationship with the consumer consciousness. 5. The most influential variables on the consumer consciousness were necessity of the education on environment, the type of school, region, and monthly allowance money in sequence. And the most influential variables on the consumer behavior were the campaign on environment in home, participation in environment preservation, cosciousness of environment, and education experience on environment in sequence. (Korean J Human Ecology 1(1) : 44∼65, 1998)

• Korean Journal of Occupational Health Nursing
• /
• v.8 no.2
• /
• pp.162-168
• /
• 1999

To investigate industrial fatigue and low back pain, the questionaire survey for subjective symptoms of fatigue and low back pain was carried out 305 nurses in general hospital in Korea. Nurses participated to this study were divided into low back pain group(LBP) and control group, according to the self-reports by written questionaires. The subjective sysptoms of fatigue comprised three groups of 10 items each, representing dullness and sleepiness(level of cerebral activation), difficulty in concentration(level of motivation) and bodily projection of fatigue. The resultant data were processed for $x^2$-test, t-test to confirm the associations. The results were as follows: 1. 28 items of fatigue subjective symptoms exept two item, 'feel thirsty and 'feel ill', were directly associated with low back pain. 2. The percentage of fatigue complaint were significantly higher in LBP group than control. 3. Of the 30 items of fatigue subjective symptoms, the highest percentage was accounted for 'legs feel heavy', 'eye strain', 'feel stiffness in the neck or the shoulders' followed by 'feel like lying', 'whole body feels tired', 'feel a pain in the low back', 'feel drowsy' and in the order of sequence. 4. The average weighted score for the first group of fatigue items(dullness and sleepiness) was the largest among three groups and was followed by the second group(difficulty in concentration) and the third group (bodily projection of fatigue) in the order of sequence, suggesting the heavier mental and shift work stress of the female workers.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.19 no.5
• /
• pp.730-736
• /
• 2009

It is more important to analysis of huge gemonics data in Bioinformatics. Here we present a novel datamining approach to predict structure and function using protein's primnary structure only. We propose not also to develope n-Block substring search algorithm in reducing enormous search space effectively in relation to feature selection, but to formulate weighted linear algorithm in a prediction of structure and function of a protein using primary structure. And we show efficient in protein domain characterization and classification by calculation weight value in determining domain association in each selected substring, and also reveal that more efficient results are acquired through claculated model score result in an inference about degree of association with each CDS(coding sequence) in domain.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.483-483
• /
• 2022

댐과 저수지의 운영 최적화를 위한 수문학적 예보는 현재 수동적인 댐 운영이 주를 이루면서 활용도가 높지 않다. 불확실한 기후변화나 기후재난 상황에서 우리 사회에 악영향을 최소화하기 위해 선제적으로 대응/대비할 수 있는 댐 운영 방안이 불가피하다. 강우량 예측 기술은 기후변화로 인해 제한적인 상황이다. 실례로, 2020년 8월에 섬진강의 댐이 극심한 집중 강우로 인해 무너지는 사태가 발생하였고 이로 인해 지역사회에 막대한 경제적 피해가 발생하였다. 선제적 댐 방류량 운영 기술은 또한 환경적인 변화로 인한 영향을 완화하기 위해 필요한 것이다. 제한적인 기상 예보 기술을 극복하고자 심화학습이나 강화학습 같은 인공지능 모델들의 활용성에 대한 연구가 시도되고 있다. 따라서 본 연구는 섬진강 댐의 시간당 수문 데이터를 이용하여 댐 운영을 위한 심화학습 모델을 개발하고 그 활용도를 평가하였다. 댐 운영을 위한 심화학습 모델로서 시계열 데이터 예측에 적합한 Long Sort Term Memory(LSTM)과 Gated Recurrent Unit(GRU) 알고리즘을 구축하고 댐 수위를 예측하였다. 분석 자료는 WAMIS에서 제공하는 2000년부터 2021년까지의 시간당 데이터를 사용하였다. 입력 데이터로서 시간당 유입량, 강우량과 방류량을, 출력 데이터로서 시간당 수위 자료를 각각 사용하였으며. 결정계수(R2 Score)를 통해 모델의 예측 성능을 평가하였다. 댐 수위 예측값 개선을 위해 하이퍼파라미터의 '최적값'이 존재하는 범위를 줄여나가는 하이퍼파라미터 최적화를 두 가지 방법으로 진행하였다. 첫 번째 방법은 수동적 탐색(Manual Search) 방법으로 Sequence Length를 24, 48, 72시간, Hidden Layer를 1, 3, 5개로 설정하여 하이퍼파라미터의 조합에 따른 LSTM와 GRU의 민감도를 평가하였다. 두 번째 방법은 Grid Search로 최적의 하이퍼파라미터를 찾았다. 이 두가지 방법에서는 같은 하이퍼파라미터 안에서 GRU가 LSTM에 비해 더 높은 예측 정확도를 보였고 Sequence Length가 높을수록 정확도가 높아지는 경향을 보였다. Manual Search 방법의 경우 R2가 최대 0.72의 정확도를 보였고 Grid Search 방법의 경우 R2가 0.79의 정확도를 보였다. 본 연구 결과는 가뭄과 홍수와 같은 물 재해에 사전 대응하고 기후변화에 적응할 수 있는 댐 운영 개선에 도움을 줄 수 있을 것으로 판단된다.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.6
• /
• pp.245-254
• /
• 2022

Proteins are the basic unit of all life activities, and understanding them is essential for studying life phenomena. Since the emergence of the machine learning methodology using artificial neural networks, many researchers have tried to predict the function of proteins using only protein sequences. Many combinations of deep learning models have been reported to academia, but the methods are different and there is no formal methodology, and they are tailored to different data, so there has never been a direct comparative analysis of which algorithms are more suitable for handling protein data. In this paper, the single model performance of each algorithm was compared and evaluated based on accuracy and speed by applying the same data to CNN, LSTM, and GRU models, which are the most frequently used representative algorithms in the convergence research field of predicting protein functions, and the final evaluation scale is presented as Micro Precision, Recall, and F1-score. The combined models CNN-LSTM and CNN-GRU models also were evaluated in the same way. Through this study, it was confirmed that the performance of LSTM as a single model is good in simple classification problems, overlapping CNN was suitable as a single model in complex classification problems, and the CNN-LSTM was relatively better as a combination model.

• Journal of fish pathology
• /
• v.37 no.1
• /
• pp.25-36
• /
• 2024

In this study, we analyzed the phylogenetic classification, epitope prediction, and pathogenicity of red sea bream iridovirus (RSIV) isolated from rock bream between 2019 and 2023. Phylogenetics based on genes encoding MCP and ATPase indicated that all five RSIV isolates belonged to RSIV subtype II. The deduced amino acid sequence of the MCP for the amplicons (1362 bp) obtained from RSIV isolates had a length of 453 amino acids. Among these, the amino acid sequences of the RSIV-19, 21, 22, and 23 isolates showed 100% identity, while the RSIV-20 isolate showed 99.78% identity with one residue difference at position 306. As a result of antigenicity analysis based on amino acid sequence, the antigenicity score of the RSIV-20 isolate was 0.6386 and the other RSIV isolates were 0.6365. Additionally, the prediction of their antigenic determinants resulted in a total of 17 identical antigenic plots. When each RSIV was inoculated into rock bream, no significant differences were observed with 100% cumulative mortality in all groups. This study provides data on the potential for genetic variation of RSIV isolated in the same marine area over the past five years, and the antigenicity and pathogenicity results of each isolate are expected to be useful information for selecting future vaccine strains.

• Journal of Animal Science and Technology
• /
• v.46 no.4
• /
• pp.537-546
• /
• 2004

Adiponectin is adipocyte complement-related protein which is highly specialized to play important roles in metabolic and honnonal processes. This protein, called GBP-28, AdipoQ, and Acrp30, is encoded by the adipose most abundant gene transcript 1 (APM1) which locates on human chromosome 3q27 and mouse chromosome 16. In order to determine chromosomal localization of the porcine APM1, we carried out PCR analysis using somatic cell hybrid panel as well as porcine whole genome radiation hybrid (RH) panel. The result showed that the porcine APM1 located on chromosome 13q41 or 13q46-49. These locations were further investigated with the two point analysis of RH panel, revealed the most significant linked marker (LOD score 20.29) being SIAT1 (8 cRs away), where the fat-related QTL located. From the SSCP analysis of APM1 using 8 pig breeds, two distinct SSCP types were detected from K~ native and Korean wild pigs. The determined sequences in Korean native and Korean wild pigs showed that two nucleotide positions (T672C and C705G) were substituted. The primary sequence of the porcine APM1 has 79 to 87% identity with those of human, mouse, and bovine APM1. The domain structures of the porcine APM1 such as signal sequence, hypervariable region, collagenous region. and globular domain are also similar to those of mammalian genes.

• Asian-Australasian Journal of Animal Sciences
• /
• v.26 no.8
• /
• pp.1065-1071
• /
• 2013

The swine leukocyte antigen class II molecules are possibly associated with the induction of protective immunity. The study described here was to investigate the relationship between polymorphisms in exon 2 of the swine DQA gene and piglet diarrhea. This study was carried out on 425 suckling piglets from three purebred pig strains (Large White, Landrace and Duroc). The genetic diversity of exon 2 in swine DQA was detected by PCR-SSCP and sequencing analysis, eight unique SSCP patterns (AB, BB, BC, CC, CD, BD, BE and DD) representing five specific allele (A to E) sequences were detected. Sequence analysis revealed 21 nucleotide variable sites and resulting in 12 amino acid substitutions in the populations. A moderate level polymorphism and significant deviations from Hardy-Weinberg equilibrium of the genotypes distribution were observed in the populations (p<0.01). The association analysis indicated that there was a statistically significant difference in the score of piglet diarrhea between different genotypes, individuals with genotype CC showed a lower diarrhea score than genotypes AB ($0.98{\pm}0.09$), BB ($0.85{\pm}0.77$) and BC ($1.25{\pm}0.23$) (p<0.05), and significantly low than genotype BE ($1.19{\pm}0.19$) (p<0.01), CC genotype may be a most resistance genotype for piglet diarrhea.

• Journal of Korea Multimedia Society
• /
• v.22 no.11
• /
• pp.1223-1232
• /
• 2019

The injection is the input method of the image feature vector from the encoder to the decoder. Since the image feature vector contains object details such as color and texture, it is essential to generate image captions. However, the bidirectional decoder model using the existing injection method only inputs the image feature vector in the first step, so image feature vectors of the backward sequence are vanishing. This problem makes it difficult to describe the context in detail. Therefore, in this paper, we propose the parallel injection method to improve the description performance of image captions. The proposed Injection method fuses all embeddings and image vectors to preserve the context. Also, We optimize our image caption model with Bidirectional Gated Recurrent Unit (Bi-GRU) to reduce the amount of computation of the decoder. To validate the proposed model, experiments were conducted with a certified image caption dataset, demonstrating excellence in comparison with the latest models using BLEU and METEOR scores. The proposed model improved the BLEU score up to 20.2 points and the METEOR score up to 3.65 points compared to the existing caption model.

• Asian-Australasian Journal of Animal Sciences
• /
• v.20 no.3
• /
• pp.328-333
• /
• 2007

Carboxypeptidase E (CPE) plays an important role in the regulation of the body fat content. Therefore, it has been suggested as candidate gene for traits related to meat quality in beef cattle. This study was conducted to identify single nucleotide polymorphisms (SNPs) in the CPE gene and to investigate association of SNP marker with carcass and meat quality traits in Korean cattle. Three SNPs were identified in the intron 4 (A309G SNP and C445T SNP) and exon 5 (C601T SNP) of the CPE gene by sequence analyses of CPE cDNA and genomic DNA samples. The sequences have been deposited in GenBank database with accession numbers AY970664 and AY970663. Genotyping of the gene-specific SNP marker was carried out using the PCR-RFLP with restriction enzymes DdeI for C445T SNP and NlaIII for C601T SNP. The frequencies of C and T alleles were 0.43 and 0.57 for C445T SNP and 0.42 and 0.58 for C601T SNP, respectively. Statistical analysis indicated that the C445T SNP showed a significant effect (p<0.05) on marbling score (MS) and breeding value of backfat thickness (BF-EBV), respectively. Animals with the CT genotype showed higher marbling score and backfat thickness than those with the TT genotype. This marker also showed a significant dominance effect for the MS and BF-EBV (p<0.05). However, no significant associations were observed between C601T SNP genotypes and all traits examined. The results suggest that the CPE gene may be used as a marker for carcass traits in Korean cattle.