• Korean Circulation Journal
• /
• v.53 no.11
• /
• pp.758-771
• /
• 2023

Background and Objectives: Paroxysmal atrial fibrillation (AF) is a major potential cause of embolic stroke of undetermined source (ESUS). However, identifying AF remains challenging because it occurs sporadically. Deep learning could be used to identify hidden AF based on the sinus rhythm (SR) electrocardiogram (ECG). We combined known AF risk factors and developed a deep learning algorithm (DLA) for predicting AF to optimize diagnostic performance in ESUS patients. Methods: A DLA was developed to identify AF using SR 12-lead ECG with the database consisting of AF patients and non-AF patients. The accuracy of the DLA was validated in 221 ESUS patients who underwent insertable cardiac monitor (ICM) insertion to identify AF. Results: A total of 44,085 ECGs from 12,666 patient were used for developing the DLA. The internal validation of the DLA revealed 0.862 (95% confidence interval, 0.850-0.873) area under the curve (AUC) in the receiver operating curve analysis. In external validation data from 221 ESUS patients, the diagnostic accuracy of DLA and AUC were 0.811 and 0.827, respectively, and DLA outperformed conventional predictive models, including CHARGE-AF, C2HEST, and HATCH. The combined model, comprising atrial ectopic burden, left atrial diameter and the DLA, showed excellent performance in AF prediction with AUC of 0.906. Conclusions: The DLA accurately identified paroxysmal AF using 12-lead SR ECG in patients with ESUS and outperformed the conventional models. The DLA model along with the traditional AF risk factors could be a useful tool to identify paroxysmal AF in ESUS patients.

• Journal of Korea Multimedia Society
• /
• v.25 no.2
• /
• pp.432-439
• /
• 2022

This paper introduces deliberative behavior tree using utility theory. The proposed approach combine the strengths of behavior trees and utility theory to implement complex behavior of AI agents in an easier and more concise way. To achieve this goal, we devised and implemented three types of additional behavior tree nodes, which evaluate utility values of its own node or its subtree while traversing and selecting its child nodes based on the evaluated values. In order to validate our approach, we implemented a sample scenario using conventional behavior tree and our proposed deliberative tree respectively. And then we compared and analyzed the simulation results.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.379-384
• /
• 2022

관계 추출은 두 개체 간의 관계를 식별하는 작업이며, 비정형 텍스트를 구조화시키는 역할을 하는 작업 중 하나이다. 현재 관계 추출에서 다양한 모델에 대한 연구들이 진행되고 있지만, 한국어 관계 추출 모델에 대한 연구는 영어에 비해 부족하다. 따라서 본 논문에서는 NE(Named Entity)태그 정보가 반영된 TEM(Typed Entity Marker)과 의존 구문 그래프를 이용한 한국어 관계 추출 모델을 제안한다. 모델의 학습과 평가 말뭉치는 KLUE에서 제공하는 관계 추출 학습 말뭉치를 사용하였다. 실험 결과 제안 모델이 68.57%의 F1 점수로 실험 모델 중 가장 높은 성능을 보여 NE태그와 구문 정보가 관계 추출 성능을 향상시킬 수 있음을 보였다.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.528-531
• /
• 2021

띄어쓰기에 대한 오류는 한국어 처리 전반에 영향을 주므로 자동 띄어쓰기는 필수적인 요소이다. 글쓴이의 대부분은 띄어쓰기 오류를 범하지 않으므로 글쓴이의 의도가 띄어쓰기 시스템에 반영되어야 한다. 그러나 대부분의 자동 띄어쓰기 시스템은 모든 띄어쓰기 정보를 제거하고 새로이 공백문자를 추가하는 방법으로 띄어쓰기를 수행한다. 이런 문제를 완화하기 위해서 본 논문에서는 기계학습에서 글쓴이의 의도가 반영된 자질을 추가하는 방법을 제안한다. 실험을 위해서 CRFs(Conditional Random Fields)를 사용하여 기존 시스템과 사용자의 의도를 반영한 띄어쓰기 시스템과의 성능을 비교하고 분석한다.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.536-539
• /
• 2021

본 논문은 한국어에서 형태소 복원을 위한 새로운 방법을 제안한다. 일반적으로 기계학습 기반 형태소 분석에서 형태소 복원은 기분석 사전과 약간의 경험규칙을 이용한다. 이와 같은 방법은 모호성을 해결하기 위해 사전에 모든 정보를 저장하는 것이 불가능할 뿐 아니라 단음절 이형태의 모호성을 해결할 수 없을 것이다. 이러한 문제를 완화하기 위해 본 논문에서는 생성된 모호성을 Viterbi 알고리즘을 이용해서 해소한다. 본 논문의 형태소 복원 과정은 기본적으로 기분석 사전과 약간의 경험규칙을 이용하여 형태소 복원 후보를 찾고 여러 후보가 있을 경우(모호성의 생성), 그 결과를 Viterbi 알고리즘으로 이형태를 결정한다. 실험을 위해 모두의 말뭉치(형태 분석)를 사용하고, 평가는 NER 방식으로 평가한다. 그 결과 품사 부착에 대해 96.28%정도의 성능을 보여주었다.

• Journal of Science Education
• /
• v.47 no.1
• /
• pp.37-51
• /
• 2023

This study explored the factors influencing major satisfaction and academic satisfaction of science education major students at the College of Education using machine learning models, random forest, gradient boosting model, and SHAP. Analysis results showed that the performance of the gradient boosting model was better than that of the random forest, but the difference was not large. Factors influencing major satisfaction include 'satisfaction with science teachers in high school corresponding to the subject of one's major', 'motivation for teaching job', and 'age'. Through the SHAP value, the influence of variables was identified, and the results were derived for the group as a whole and for individual analysis. The comprehensive and individual results could be complementary with each other. Based on the research results, implications for ways to support pre-service science teachers' major and academic satisfaction were proposed.

• Journal of Korean Society for Atmospheric Environment
• /
• v.15 no.4
• /
• pp.393-402
• /
• 1999

Precipitation samples were collected by the wet-only sampling method in Iksan in the northwest of Chonbuk from March 1995 to February 1997. These samples were analyzed for the concentration of ion components, in addition to pH and electrical conductivity. The annual mean pH of precipitation was 4.8 and the seasonal trend of pH was shown to be low in Fall and Winter(4.5), middle-ranged in Spring(4.7) and high in Summer(5.0). The frequency of pH below 5.6 was about 71%. The seasonal pattern of pH frequency was found to be different in each season. In the case of the pH less than 5.0, the frequency was higher in Spring, Fall and Winter than in Summer, especially higher in Fall than in other seasons. The concentrations of analysed ions showed a pronounced seasonal pattern. However, major ion species for all seasons were $NH^+_4,;Ca^{2+};and;Na^+$ among cations and $SO^{2-}_4,;Cl^-;and;NO^-_3$ among anions. The major acidifying species appeared to be $nss-SO^{2-}_4;and;NO^-_3$, and the main bases responsible for the neutralization of precipitation acidity were $nss-Ca^{2+};and;NH^+_4$. The potential acidity of precipitation, pAi, was found to be between 3.0 and 5.0 for total samples, while the measured pH was approximately between 3.9 and 7.8. The seasonal trend of pAi showed a decreasing order: Summer (4.3), Winter(4.0), Spring and Fall(3.8). During the Fall, both pAi and pH were especially very low, which indicated that during this period the potential acidity of precipitation was high but the neutralizing capacity was low. For Spring, pAi was very low but pH was slightly high. This was likely due to the large amount of $CaCO_3$ in the soil particles transported over a long range from the Chinese continent that were incorporated into the precipitation, and then neutralized the acidifying species with its high concentraton.

• Journal of The Korean Association of Information Education
• /
• v.25 no.2
• /
• pp.301-308
• /
• 2021

In line with the social changes due to the Fourth Industrial Revolution, foreign countries are strengthening information education for the future of education. This study analyzed the current status of software (SW) and artificial intelligence (AI) education among different types of information education programs in elementary schools of major foreign countries, and compared them with the education provided in Korea. Compared to major foreign countries, Korea allocated very little time for software education in elementary schools, making it difficult to sufficiently cover all areas of the curriculum achievement standards. In addition, other countries recognized the importance of artificial intelligence, an important technology of the Fourth Industrial Revolution, and were providing artificial intelligence education on the basis of software education at the national level. The Korean government is also planning on providing the education at national level, but it was identified that the information education of elementary schools have many problematic issues. This study emphasized the need to establish an information curriculum for elementary schools as a way to address these issues.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2022.06a
• /
• pp.68-71
• /
• 2022

항로표지는 해상에 독립적으로 암초 위나 줄에 의해 떠 있는 형태로 존재하며, 선박들의 안전 운행에 필요한 다양한 정보를 제공하는 역할을 수행한다. 이러한 항로표지의 설치 및 동작 형태는 풍랑에 따라 기기의 위치가 가변적으로 변하게 된다. 따라서 기기의 위치가 급격하게 변했을 때, 항로표지 기기 내에도 영향을 받는다면 지방청의 항로표지 데이터 수신이 낮아질 것이라고 가설 설정했다. 본 논문에서는 기상특보에 따른 시간적 기준으로 구간을 나누어 풍랑과 항로표지 데이터 수신 간의 상관관계가 있는지 연구를 진행하였다. 연구 결과 풍랑이 거세질수록 평균 데이터 수집량이 감소하는 것으로 데이터 수신 강도의 영향을 줄 수 있음을 확인하였다. 이번 연구를 통해 풍랑에 대비한 항로표지 데이터의 개선이 필요하며, 선박의 안전과 관련된 만큼 정밀한 개선을 요한다.

• Journal of Korea Multimedia Society
• /
• v.24 no.8
• /
• pp.979-987
• /
• 2021

Multiple Sclerosis (MS) can be early diagnosed by detecting lesions in brain magnetic resonance images (MRI). Unsupervised anomaly detection methods based on autoencoder have been recently proposed for automated detection of MS lesions. However, these autoencoder-based methods were developed only for 2D images (e.g. 2D cross-sectional slices) of MRI, so do not utilize the full 3D information of MRI. In this paper, therefore, we propose a novel 3D autoencoder-based framework for detection of the lesion volume of MS in MRI. We first define a 3D convolutional neural network (CNN) for full MRI volumes, and build each encoder and decoder layer of the 3D autoencoder based on 3D CNN. We also add a skip connection between the encoder and decoder layer for effective data reconstruction. In the experimental results, we compare the 3D autoencoder-based method with the 2D autoencoder models using the training datasets of 80 healthy subjects from the Human Connectome Project (HCP) and the testing datasets of 25 MS patients from the Longitudinal multiple sclerosis lesion segmentation challenge, and show that the proposed method achieves superior performance in prediction of MS lesion by up to 15%.