• 말소리와 음성과학
• /
• 제13권1호
• /
• pp.73-79
• /
• 2021

본 연구에서는 노화에 따른 발화 시 입술 움직임의 변화를 알아보고자 하였다. 연구대상으로 평균 69세의 노인 여성 15명과 평균 22세의 젊은 여성 15명을 선정하였다. 입술 움직임을 측정하기 위해 이중모음 발화시 입술 움직임을 녹화하여 스틸 이미지로 저장한 다음 입술의 움직임이 최소인 부분과 최대한 길이를 영상분석 소프트웨어를 이용하여 pixel 단위로 수작업으로 분석하여 비교하였다. 임상적 활용성을 위해 자동화 알고리즘을 적용하여 소프트웨어를 제작했으며 수작업의 결과와 비교하였다. 연구결과 노년층의 경우 청년층에 비해 이중모음 과제에서 입술의 가로 및 세로의 길이 범위가 작은 것을 알 수 있었다. 수작업과 자동화 방법의 상관관계를 측정한 결과 강한 정적 상관관계가 나타나 두 방법 모두 입술 윤곽 추출 시 유용함을 알 수 있었다. 이상의 결과를 바탕으로 노화가 진행됨에 따라 발화 시 입술의 범위가 작아지는 것을 알 수 있었다. 따라서 노화가 진행되기 전 간단하게 입술의 움직임을 측정하여 본인의 상태를 모니터링하고 입술 범위를 유지할 수 있는 운동을 실시한다면 노화로 인한 발음 문제를 예방할 수 있을 것이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2049-2068
• /
• 2021

AI combined with NLP techniques has promoted the use of Virtual Assistants and have made people rely on them for many diverse uses. Conversational Agents are the most promising technique that assists computer users through their operation. An important challenge in developing Conversational Agents globally is transferring the groundbreaking expertise obtained in English to other languages. AI is making it possible to transfer this learning. There is a dire need to develop systems that understand secular languages. One such difficult language is Hindi, which is the fourth most spoken language in the world. Semantic similarity is an important part of Natural Language Processing, which involves applications such as ontology learning and information extraction, for developing conversational agents. Most of the research is concentrated on English and other European languages. This paper presents a Corpus-based word semantic similarity measure for Hindi. An experiment involving the translation of the English benchmark dataset to Hindi is performed, investigating the incorporation of the corpus, with human and machine similarity ratings. A significant correlation to the human intuition and the algorithm ratings has been calculated for analyzing the accuracy of the proposed similarity measures. The method can be adapted in various applications of word semantic similarity or module for any other language.

• 정보보호학회논문지
• /
• 제13권1호
• /
• pp.139-146
• /
• 2003

본 논문은 과포화 다변수 방정식을 이용하여 LILI-128 스트림 암호를 분석한다. LILI-128 암호$^{[8]}$ 는 128비트 키를 가진 선형귀환 쉬프트 레지스터 기반의 스트림 암호로 구조를 살펴보면 크게 “CLOCK CONTROL” 부분과 “DATA GENERATION” 부분으로 나뉘어진다. 분석 방법은 “DATA CENERATION” 부분에 사용되는 함수 \ulcorne $r^{d}$ 의 대수적 차수가 높지 못하다는 성질을 이용한다. 간략히 설명하면 차수(K)가 6차인 다변수 방정식을 많이 얻을 수 있고, 이를 7차 (D)의 다변수 방정식으로 확장하여 주어진 변수보다 많은 연립방정식을 얻어 그 해를 구하는 XL 알고리즘을 통해 전수조사보다 빠르게 키정보를 찾을 수 있다. 결과 중 가장 좋은 것은 출력 키수열 2$^{26.3}$비트를 가지고 2$^{110.7}$ CPU 시간을 통해 128비트 키정보를 얻는 것이다.다.

• 융합정보논문지
• /
• 제11권7호
• /
• pp.31-38
• /
• 2021

본 논문은 감성 분석에 사용되는 RNN 모델의 최적화를 얻기 위한 성능분석을 위하여 하이퍼파라미터 튜닝에 따른 손실과 정확도의 추이를 관찰하여 모델과의 상관관계를 연구하였다. 연구 방법으로는 시퀀셜데이터를 처리하는데 가장 최적화된 LSTM과 Embedding layer로 히든레이어를 구성한 후, LSTM의 Unit과 Batch Size, Embedding Size를 튜닝하여 각각의 모델에 대한 손실과 정확도를 측정하였다. 측정 결과, 손실은 41.9%, 정확도는 11.4%의 차이를 나타내었고, 최적화 모델의 변화추이는 지속적으로 안정적인 그래프를 보여 하이퍼파라미터의 튜닝이 모델에 지대한 영향을 미침을 확인하였다. 또한 3가지 하이퍼파라미터 중 Embedding Size의 결정이 모델에 가장 큰 영향을 미침을 확인하였다. 향후 이 연구를 지속적으로 이어나가 모델이 최적의 하이퍼파라미터를 직접 찾아낼 수 있는 알고리즘에 대한 연구를 지속적으로 이어나갈 것이다.

• 대한의용생체공학회:의공학회지
• /
• 제42권3호
• /
• pp.100-106
• /
• 2021

Measuring Intima-media thickness (IMT) with ultrasound images can help early detection of coronary artery disease. As a result, numerous machine learning studies have been conducted to measure IMT. However, most of these studies require several steps of pre-treatment to extract the boundary, and some require manual intervention, so they are not suitable for on-site treatment in urgent situations. in this paper, we propose to use deep learning networks U-Net, Attention U-Net, and Pretrained U-Net to automatically segment the intima-media complex. This study also applied the HE, HS, and CLAHE preprocessing technique to wireless portable ultrasound diagnostic device images. As a result, The average dice coefficient of HE applied Models is 71% and CLAHE applied Models is 70%, while the HS applied Models have improved as 72% dice coefficient. Among them, Pretrained U-Net showed the highest performance with an average of 74%. When comparing this with the mean value of IMT measured by Conventional wired ultrasound equipment, the highest correlation coefficient value was shown in the HS applied pretrained U-Net.

• Archives of Plastic Surgery
• /
• 제48권3호
• /
• pp.269-277
• /
• 2021

Background The Sunnybrook facial grading scale is a comprehensive scale for the evaluation of facial paralysis patients. Its results greatly depend on subjective input. This study aimed to develop and validate an automated Sunnybrook facial grading scale (SBface) to more objectively assess disfigurement due to facial paralysis. Methods An application compatible with iOS version 11.0 and up was developed. The software automatically detected facial features in standardized photographs and generated scores following the Sunnybrook facial grading scale. Photographic data from 30 unilateral facial paralysis patients were randomly sampled for validation. Intrarater reliability was tested by conducting two identical tests at a 2-week interval. Interrater reliability was tested between the software and three facial nerve clinicians. Results A beta version of the SBface application was tested. Intrarater reliability showed excellent congruence between the two tests. Moderate to strong positive correlations were found between the software and an otolaryngologist, including the total scores of the three individual software domains and composite scores. However, 74.4% (29/39) of the subdomain items showed low to zero correlation with the human raters (κ<0.2). The correlations between the human raters showed good congruence for most of the total and composite scores, with 10.3% (4/39) of the subdomain items failing to correspond (κ<0.2). Conclusions The SBface application is efficient and accurate for evaluating the degree of facial paralysis based on the Sunnybrook facial grading scale. However, correlations of the software-derived results with those of human raters are limited by the software algorithm and the raters' inconsistency.

• 한국간호교육학회지
• /
• 제27권2호
• /
• pp.197-209
• /
• 2021

Purpose: The purpose of this study is to evaluate the validity and reliability of the Korean Bereavement Care Confidence Scale (K-BCCS). Methods: The Perinatal Bereavement Care Confidence Scale (PBCCS) was translated into Korean according to an algorithm of cultural adaptation process and excluded six items which were specific to perinatal bereavement. A total of 229 clinical nurses participated in the study. Construct validity, convergent validity, discriminant validity, and group comparison validity were evaluated, and Cronbach's α was calculated to estimate the reliability of the K-BCCS. Results: The K-BCCS consisted of 31 items in 7 factors, including knowledge and skills for bereavement care (12 items), organizational support (6 items), awareness of the needs (3 items), interpersonal skills (3 items), workload influence (2 items), continuous education (2 items), and understanding the grief process (3 items). The factor loading of 31 items within the 7 factors ranged from .60 to .86. For the convergent validity, the construct reliability (CR) ranged from .74 to .94, and the average variance extracted (AVE) ranged from .49 to .73, which is considered acceptable. The discriminant validity showed that the AVEs of the subscales were greater than the square of the correlation coefficient r. The nurses who had experience providing bereavement care (t=4.94, p<.001) or had received bereavement education (t=6.64, p<.001) showed higher K-BCCS values those without experience. The Cronbach's α of 31 items was .93 and ranged from .60 to .94 per subscale. Conclusion: The K-BCCS is a valid and reliable tool for evaluating nurses' confidence in bereavement care.

• Composite Materials and Engineering
• /
• 제3권3호
• /
• pp.201-220
• /
• 2021

In this study, potential of three machine learning techniques i.e., M5P, Support vector machines and Gaussian processes were evaluated to find the best algorithm for the prediction of flexural strength of concrete mix with steel fibre. The study comprises the comparison of results obtained from above-said techniques for given dataset. The dataset consists of 124 observations from past research studies and this dataset is randomly divided into two subsets namely training and testing datasets with (70-30)% proportion by weight. Cement, fine aggregates, coarse aggregates, water, super plasticizer/ high-range water reducer, steel fibre, fibre length and curing days were taken as input parameters whereas flexural strength of the concrete mix was taken as the output parameter. Performance of the techniques was checked by statistic evaluation parameters. Results show that the Gaussian process technique works better than other techniques with its minimum error bandwidth. Statistical analysis shows that the Gaussian process predicts better results with higher coefficient of correlation value (0.9138) and minimum mean absolute error (1.2954) and Root mean square error value (1.9672). Sensitivity analysis proves that steel fibre is the significant parameter among other parameters to predict the flexural strength of concrete mix. According to the shape of the fibre, the mixed type performs better for this data than the hooked shape of the steel fibre, which has a higher CC of 0.9649, which shows that the shape of fibers do effect the flexural strength of the concrete. However, the intricacy of the mixed fibres needs further investigations. For future mixes, the most favorable range for the increase in flexural strength of concrete mix found to be (1-3)%.

• 가정∙방문간호학회지
• /
• 제29권3호
• /
• pp.278-287
• /
• 2022

Purpose: To analyze Korean nurse-related channels and video titles on YouTube, the world's largest online video sharing and social media platform, to clarify public opinion and image of nurses. We seek utilization strategies and measures through current status analysis. Methods: Data is collected by crawling video information related to Korean nurses, and correlation is analyzed with frequent word analysis and keyword network analysis. Results: Through the YouTube algorithm, 2,273 videos of 'Nurse' were analyzed in order of recent views, relevance, and rating, and 2,912 videos searched for with the keyword 'Nurse + Hospital, COVID-19, Awareness, University, National Examination' were analyzed. Numerous videos were uploaded, and nursing work that was uploaded in the form of a vlog recorded a high number of views. Conclusion: We could see if the YouTube video shows images of nurses. It has been confirmed that various information is being exchanged rather than information just for promotional purposes.

• 지적과 국토정보
• /
• 제51권2호
• /
• pp.141-150
• /
• 2021

미세먼지는 사람의 건강에 많은 영향을 미치는 물질로서 이와 관련하여 다양한 연구가 이루어지고 있다. 미세먼지의 인체 영향으로 인해 서울시 모니터링 네트워크에서 측정된 과거 데이터를 활용하여 미세먼지를 예측하려는 다양한 연구가 진행되고 있다. 본 연구는 2019년 5월 서울시의 미세먼지를 중점으로 진행하였으며, 학습에 사용한 변수는 SO₂, CO, NO₂, O₃와 같은 대기오염물질 데이터를 활용하였다. 예측모델은 Adaboost에 기반하여 구축하였고, 훈련모델은 PM₁₀과 PM_2.5로 구분하였다. 에러 메트릭스를 통한 예측모델의 정확도 평가 결과로 Adaboost가 시도되었다. 대기오염물질은 초미세먼지와 더 높은 상관성을 보이는 것으로 나타났지만, 보다 효과적인 분포등급을 제시하기 위해서는 많은 양의 데이터를 학습하고, PM₁₀과 PM_2.5의 공간분포 등급을 효과적으로 예측하기 위해서 교통량 등의 추가적인 변수를 활용할 필요성이 있다고 판단된다.