• Journal of Information Processing Systems
• /
• 제14권6호
• /
• pp.1301-1306
• /
• 2018

Artificial intelligence is one of the key technologies of the Fourth Industrial Revolution. This paper introduces the diverse kinds of approaches to subjects that tackle diverse kinds of research fields such as model-based MS approach, deep neural network model, image edge detection approach, cross-layer optimization model, LSSVM approach, screen design approach, CPU-GPU hybrid approach and so on. The research on Superintelligence and superconnection for IoT and big data is also described such as 'superintelligence-based systems and infrastructures', 'superconnection-based IoT and big data systems', 'analysis of IoT-based data and big data', 'infrastructure design for IoT and big data', 'artificial intelligence applications', and 'superconnection-based IoT devices'.

• 한국BIM학회 논문집
• /
• 제12권1호
• /
• pp.1-9
• /
• 2022

Concrete formwork is a crucial component for any construction project. Artificial intelligence offers great potential to automate formwork design by offering various design options and under different criteria depending on the requirements. This study applied image segmentation in 2D formwork drawings to extract sheathing, strut and pipe support formwork elements. The proposed artificial intelligence model can recognize, classify, and extract formwork elements from 2D CAD drawing image and training and test results confirmed the model performed very well at formwork element recognition with average precision and recall better than 80%. Recognition systems for each formwork element can be implemented later to generate 3D BIM models.

• The Plant Pathology Journal
• /
• 제38권4호
• /
• pp.395-402
• /
• 2022

To predict rice blast, many machine learning methods have been proposed. As the quality and quantity of input data are essential for machine learning techniques, this study develops three artificial neural network (ANN)-based rice blast prediction models by combining two ANN models, the feed-forward neural network (FFNN) and long short-term memory, with diverse input datasets, and compares their performance. The Blast_Weathe long short-term memory r_FFNN model had the highest recall score (66.3%) for rice blast prediction. This model requires two types of input data: blast occurrence data for the last 3 years and weather data (daily maximum temperature, relative humidity, and precipitation) between January and July of the prediction year. This study showed that the performance of an ANN-based disease prediction model was improved by applying suitable machine learning techniques together with the optimization of hyperparameter tuning involving input data. Moreover, we highlight the importance of the systematic collection of long-term disease data.

• 반도체디스플레이기술학회지
• /
• 제21권4호
• /
• pp.110-115
• /
• 2022

Accurate current-voltage modeling of solar cell systems plays an important role in power prediction. Solar cells have nonlinear characteristics that are sensitive to environmental conditions such as temperature and irradiance. In this paper, the output characteristics of photovoltaic module are accurately predicted by combining the artificial neural network and physical model. In order to estimate the performance of PV module under varying environments, the artificial neural network model is trained with randomly generated temperature and irradiance data. With the use of proposed model, the current-voltage and power-voltage characteristics under real environments can be predicted with high accuracy.

• Computers and Concrete
• /
• 제31권4호
• /
• pp.315-325
• /
• 2023

Calcium leaching is one of the main deterioration factors in concrete structures contact with water, such as dams, water treatment structures, and radioactive waste structures. It causes a porous microstructure and may be coupled with various harmful factors resulting in mechanical degradation of concrete. Several numerical modeling studies focused on the calcium leaching depth prediction. However, these required a lot of cost and time for many experiments and analyses. This study presents an artificial neural network (ANN) approach to predict the leaching depth quickly and accurately. Totally 132 experimental data are collected for model training and validation. An optimal ANN model was proposed by ANN topology. Results indicate that the model can be applied to estimate the calcium leaching depth, showing the determination coefficient of 0.91. It might be used as a simulation tool for engineering problems focused on durability.

• 한국콘텐츠학회논문지
• /
• 제21권2호
• /
• pp.587-593
• /
• 2021

본 논문에서는 나날이 발전하는 대규모 서비스거부공격에 대해 고전적인 방식의 DDoS 대응시스템에서 벗어나, 4차 혁명 시대의 핵심기술 중의 하나인 인공지능 기반의 기술을 활용해 지능화된 서비스거부공격을 효율적으로 감내 할 수 있는 서비스모델 개발방안을 제안하였다. 즉, 다수의 보안장비, 웹서버로부터 수집된 다량의 데이터를 대상으로 머신러닝 인공지능 학습을 통해 서비스거부공격을 탐지하고 피해를 최소화할 수 있는 방안을 제안하였다. 특히, 인공지능기술을 활용하기 위한 모델을 개발은 일정한 트래픽 변화를 반복하며 안정적 흐름의 데이터를 전송이 이루어지다가 서비스거부공격이 발생하면 다른 양상의 데이터 흐름을 보인다는 점에 착안하여 서비스서부공격 탐지에 인공지능기술을 활용하였다. 서비스거부공격이 발생하면 확률기반의 실제 트래픽과 예측값과의 편차가 발생하기 때문에 공격성 데이터로 판단하여 대응이 가능하다. 이 논문에서는 보안장비나 서버에서 발생하는 로그를 기반으로 데이터를 분석하여 서비스거부공격 탐지모델을 설명하였다.

• 디지털융복합연구
• /
• 제16권1호
• /
• pp.231-242
• /
• 2018

병원 재원일수의 효율적 관리는 병원의 수익과 환자의 진료비 절감을 위해 매우 중요한 요소이다. 이러한 재원일수의 효율적 관리를 위해서는 병원들이 재원일수에 대해서 벤치마킹을 할 수 있도록 지원이 필요하고 재원일수 절감의 구체적인 방향을 제시해 줄 수 있는 재원일수 예측모형의 개발이 필요하다. 본 연구에서는 2013년과 2014년도 퇴원손상환자자료 중 급성뇌졸중 환자를 추출하여 분석용 자료를 만들고 인공지능을 이용하여 급성뇌졸중 환자의 재원일수 예측모형을 개발하였다. 분석용 자료는 훈련용 60%, 평가용 40%로 분류하였다. 모형개발은 전통적 통계기법인 다중회귀분석기법과 인공지능기법인 대화식 의사결정나무기법, 신경망 기법, 그리고 이들을 모두 통합한 앙상블기법을 이용하였다. 모형평가는 Root ASE(Absolute error) 지표를 이용하였는데, 다중회귀분석은 23.7, 대화식결정나무 23.7, 신경망 분석은 22.7, 앙상블은 22.7로 나타났고 이를 통하여 재원일수 예측모형 개발에 인공지능기법의 유용성이 입증되었다. 앞으로 재원일수 예측모형개발에 인공지능 기법을 보다 효율적으로 활용할 수 있는 방안에 대해서 계속적인 연구가 이루어 질 필요가 있다.

• 대한기계학회:학술대회논문집
• /
• 대한기계학회 2000년도 추계학술대회논문집A
• /
• pp.345-350
• /
• 2000

In this paper, we introduce a new artificial material model for topology optimization. The present material model, named S-shape material model, accelerates topology optimization process especially in mathematical programming. We overcome the instability and the flatness in heuristic optimization process. Numerical examples show the superiority of the proposed material.

• International Journal of Computer Science & Network Security
• /
• 제22권4호
• /
• pp.119-130
• /
• 2022

Crime is a common social problem that affects the quality of life. As the number of crimes increases, it is necessary to build a model to predict the number of crimes that may occur in a given period, identify the characteristics of a person who may commit a particular crime, and identify places where a particular crime may occur. Data privacy is the main challenge that organizations face when building this type of predictive models. Federated learning (FL) is a promising approach that overcomes data security and privacy challenges, as it enables organizations to build a machine learning model based on distributed datasets without sharing raw data or violating data privacy. In this paper, a federated long short- term memory (LSTM) model is proposed and compared with a traditional LSTM model. Proposed model is developed using TensorFlow Federated (TFF) and the Keras API to predict the number of crimes. The proposed model is applied on the Boston crime dataset. The proposed model's parameters are fine tuned to obtain minimum loss and maximum accuracy. The proposed federated LSTM model is compared with the traditional LSTM model and found that the federated LSTM model achieved lower loss, better accuracy, and higher training time than the traditional LSTM model.

• 한국지반공학회논문집
• /
• 제19권6호
• /
• pp.343-350
• /
• 2003

인공신경망(Artificial Neural Network, ANN) 해석기술을 지반공학 분야에서 활용하는 경우가 점점 다양해지고 있다. 이 연구에서는 초음파에 의해 증가된 토양수세법의 효율성을 해석하는 모델개발에 인공신경망기법을 적용하였다. 실내시험을 통하여 인공신경망을 위한 입력자료를 확보한 뒤 이를 이용하여 모델을 학습시킨 후 모델검증을 실시하였다. 해석 변수, 즉 모멘텀항, 학습률, 전이함수 종류, 은닉층 수 및 노드 수 등을 달리하여 연구를 수행하였으며 최적의 조건을 도출한 후 개발된 모델의 검증을 실시하였다. 개발된 모델의 검증결과 측정값과 예측값의 상관관계가 매우 높게 나타났으며 이를 통하여 수학적 모델 수립이 곤란한 토양수세 초음파 기법의 전반적인 고찰의 기초를 확립하였다.