• 한국표면공학회지
• /
• 제56권6호
• /
• pp.353-370
• /
• 2023

Corrosion detection and analysis is a very important topic in reducing costs and preventing disasters. Recently, image processing techniques have been widely applied to corrosion identification and analysis. In this work, we briefly introduces traditional image processing techniques and machine learning algorithms applied to detect or analyze corrosion in various fields. Recently, machine learning, especially CNN-based algorithms, have been widely applied to corrosion detection. Additionally, research on applying machine learning to region segmentation is very actively underway. The corrosion is reddish and brown in color and has a very irregular shape, so a combination of techniques that consider color and texture, various mathematical techniques, and machine learning algorithms are used to detect and analyze corrosion. We present examples of the application of traditional image processing techniques and machine learning to corrosion detection and analysis.

• 대한한의학원전학회지
• /
• 제35권1호
• /
• pp.93-102
• /
• 2022

Objectives : We propose a method for constructing a thesaurus of Traditional East Asian Medicine terminology using machine learning. Methods : We presented a method of combining the 'Automatic Step' which uses machine learning and the 'Manual Step' which is the operator's review process. By applying this method to the sample data, we constructed a simple thesaurus and examined the results. Results : Out of the 17,874 sample data, a thesaurus was constructed targeting 749 terminologies. 200 candidate groups were derived in the automatic step, from which 79 synonym groups were derived in the manual step. Conclusions : The proposed method in this study will likely save resources required in constructing a thesaurus.

• 한국전자통신학회논문지
• /
• 제18권1호
• /
• pp.85-90
• /
• 2023

보건의료분야 데이터 분석 방법론이 기존의 통계 중심의 연구방법에서 머신러닝을 이용한 예측 연구로 전환되고 있다. 본 연구에서는 다양한 머신러닝 도구들을 살펴보고, 보건의료분야에서 많이 사용하고 있는 통계 도구인 R을 빅데이터 머신러닝에 적용하기 위해 R과 스파크를 연계한 프로그래밍 모델들을 비교한다. 그리고, R을 스파크 환경에서 수행하는 SparkR을 이용한 선형회귀모델 학습의 성능을 스파크의 기본 언어인 스칼라를 이용한 모델과 비교한다. 실험 결과 SparkR을 이용할 때의 학습 수행 시간이 스칼라와 비교하여 10~20% 정도 증가하였다. 결과로 제시된 성능 저하를 감안한다면 기존의 통계분석 도구인 R을 그대로 활용 가능하다는 측면에서 SparkR의 분산 처리의 유용성을 확인하였다.

• 대한한의학원전학회지
• /
• 제36권2호
• /
• pp.23-34
• /
• 2023

Objectives : We propose a method to disassemble Traditional East Asian Medicine herbal formulas using machine learning. Methods : After creating a model using Byte Pair Encoding(BPE) and G-Score, the model was trained with training data. Afterwards, the learned model was applied to the test data, of which the results were compared with expert opinion. Results : The results acquired through the model were not significantly different from those of modern expert opinions. However, there were cases where the meaning was partially unclear, while there were cases where new knowledge could be obtained through the disassembling process. Conclusions : It is expected that disassembling herbal formulas through the proposed method in this study will help save resources required to understand complex ones.

• 한국해양공학회지
• /
• 제34권2호
• /
• pp.147-154
• /
• 2020

Underwater acoustics that is the study of the phenomenon of underwater wave propagation and its interaction with boundaries, has mainly been applied to the fields of underwater communication, target detection, marine resources, marine environment, and underwater sound sources. Based on the scientific and engineering understanding of acoustic signals/data, recent studies combining traditional and data-driven machine learning methods have shown continuous progress. Machine learning, represented by deep learning, has shown unprecedented success in a variety of fields, owing to big data, graphical processor unit computing, and advances in algorithms. Although machine learning has not yet been implemented in every single field of underwater acoustics, it will be used more actively in the future in line with the ongoing development and overwhelming achievements of this method. To understand the research trends of machine learning applications in underwater acoustics, the general theoretical background of several related machine learning techniques is introduced in this paper.

• Computers and Concrete
• /
• 제33권3호
• /
• pp.285-299
• /
• 2024

Recently, many engineering computations have realized their digital transformation to Machine Learning (ML)-based systems. Predicting the behavior of a structure, which is mainly computed with structural analysis software, is an essential step before construction for efficient structural analysis. Especially in the seismic-based design procedure of the structures, predicting the lateral load capacity of reinforced concrete (RC) columns is a vital factor. In this study, a novel ML-based model is proposed to predict the maximum lateral load capacity of RC columns under varying axial loads or cyclic loadings. The proposed model is generated with a Deep Neural Network (DNN) and compared with traditional ML techniques as well as a popular commercial structural analysis software. In the design and test phases of the proposed model, 319 columns with rectangular and square cross-sections are incorporated. In this study, 33 parameters are used to predict the maximum lateral load capacity of each RC column. While some traditional ML techniques perform better prediction than the compared commercial software, the proposed DNN model provides the best prediction results within the analysis. The experimental results reveal the fact that the performance of the proposed DNN model can definitely be used for other engineering purposes as well.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2019년도 춘계학술대회
• /
• pp.155-156
• /
• 2019

Classification and Regression Tree (CART), SVM (Support Vector Machine) 및 k-nearest neighbor classification (kNN)과 같은 기존 기계 학습 기반 감정 분석 방법은 정확성이 떨어졌습니다. 본 논문에서는 개선 된 kNN 분류 방법을 제안한다. 개선 된 방법 및 데이터 정규화를 통해 정확성 향상의 목적이 달성됩니다. 그 후, 3 가지 분류 알고리즘과 개선 된 알고리즘을 실험 데이터에 기초하여 비교 하였다.

• 한국콘텐츠학회논문지
• /
• 제22권2호
• /
• pp.545-558
• /
• 2022

본 연구의 목적은 전통적인 통계과 기계학습(Machine Learning)을 통해 중국 문화산업 기업의 재무적 곤경을 정확하게 예측하는 분석 모형을 탐색하는 데 있다. 예측모형을 구축하기 위하여 중국 128개 문화산업상장 기업의 데이터를 수집하였다. 25개 설명변수로 이뤄진 데이터베이스를 토대로 판별분석과 로지스틱 회귀(Logistic) 등 전통적인 통계 방법과 서포트 벡터 기계(SVM), 결정 트리(Decision Tree)와 랜덤 포레스트(Random Forest) 등 기계학습을 이용한 예측모형을 구축하고 각 모형의 성능 평가를 위해 Python 소프트웨어를 사용한다. 분석 결과, 예측 성능이 가장 좋은 모형은 랜덤 포레스트(Random Forest) 모형으로 95%의 정확도를 보였다. 그 다음은 서포트 벡터 기계(SVM) 모형으로 93%의 정확도를 보였다. 그 다음은 결정 트리(Decision Tree) 모형으로 92%의 정확도를 보였다. 그 다음은 판정분석 모형으로 89%의 정확도를 보였다. 예측 효과가 가장 낮은 모형은 로지스틱 회귀(Logistic) 모형으로 88%의 정확도를 보였다. 이는 중국 문화산업 기업의 재무적 곤경을 예측하면서 기계학습 모형이 전통적인 통계 모형보다 더 좋은 예측 효과를 얻을 수 있음을 설명한다.

• 전기전자학회논문지
• /
• 제22권4호
• /
• pp.948-952
• /
• 2018

Haze removal is an object of scientific desire due to its various practical applications. Existing algorithms are founded upon histogram equalization, contrast maximization, or the growing trend of applying machine learning in image processing. Since machine learning-based algorithms solve problems based on the data, they usually perform better than those based on traditional image processing/computer vision techniques. However, to achieve such a high performance, one of the requisites is a large and reliable training database, which seems to be unattainable owing to the complexity of real hazy and haze-free images acquisition. As a result, researchers are currently using the synthetic database, obtained by introducing the synthetic haze drawn from the standard uniform distribution into the clear images. In this paper, we propose the enhanced equidistribution, improving upon our previous study on equidistribution, and use it to make a new database for training machine learning-based haze removal algorithms. A large number of experiments verify the effectiveness of our proposed methodology.

• 전자통신동향분석
• /
• 제38권5호
• /
• pp.51-60
• /
• 2023

Quantum machine learning (QML) is an area of quantum computing that leverages its principles to develop machine learning algorithms and techniques. QML is aimed at combining traditional machine learning with the capabilities of quantum computing to devise approaches for problem solving and (big) data processing. Nevertheless, QML is in its early stage of the research and development. Thus, more theoretical studies are needed to understand whether a significant quantum speedup can be achieved compared with classical machine learning. If this is the case, the underlying physical principles may be explained. First, fundamental concepts and elements of QML should be established. We describe the inception and development of QML, highlighting essential quantum computing algorithms that are integral to QML. The advent of the noisy intermediate-scale quantum era and Google's demonstration of quantum supremacy are then addressed. Finally, we briefly discuss research prospects for QML.