• 지능정보연구
• /
• 제5권1호
• /
• pp.103-123
• /
• 1999

Input filtering as a preprocessing method is so much crucial to get good performance in time series forecasting. There are a few preprocessing methods (i.e. ARMA outputs as time domain filters, and Fourier transform or wavelet transform as time-frequency domain filters) for handling time series. Specially, the time-frequency domain filters describe the fractal structure of financial markets better than the time domain filters due to theoretically additional frequency information. Therefore, we, first of all, try to describe and analyze specially some issues on the effectiveness of different filtering methods from viewpoint of the performance of a neural network based forecasting. And then we discuss about neural network model architecture issues, for example, what type of neural network learning architecture is selected for our time series forecasting, and what input size should be applied to a model. In this study an input selection problem is limited to a size selection of the lagged input variables. To solve this problem, we simulate on analyzing and comparing a few neural networks having different model architecture and also use an embedding dimension measure as chaotic time series analysis or nonlinear dynamic analysis to reduce the dimensionality (i.e. the size of time delayed input variables) of the models. Throughout our study, experiments for integration methods of joint time-frequency analysis and neural network techniques are applied to a case study of daily Korean won / U. S dollar exchange returns and finally we suggest an integration framework for future research from our experimental results.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권4호
• /
• pp.2060-2077
• /
• 2019

Recently, mobile healthcare services have attracted significant attention because of the emerging development and supply of diverse wearable devices. Smartwatches and health bands are the most common type of mobile-based wearable devices and their market size is increasing considerably. However, simple value comparisons based on accumulated data have revealed certain problems, such as the standardized nature of health management and the lack of personalized health management service models. The convergence of information technology (IT) and biotechnology (BT) has shifted the medical paradigm from continuous health management and disease prevention to the development of a system that can be used to provide ground-based medical services regardless of the user's location. Moreover, the IT-BT convergence has necessitated the development of lifestyle improvement models and services that utilize big data analysis and machine learning to provide mobile healthcare-based personal health management and disease prevention information. Users' health data, which are specific as they change over time, are collected by different means according to the users' lifestyle and surrounding circumstances. In this paper, we propose a prediction model of user physical activity that uses data characteristics-based long short-term memory (DC-LSTM) recurrent neural networks (RNNs). To provide personalized services, the characteristics and surrounding circumstances of data collectable from mobile host devices were considered in the selection of variables for the model. The data characteristics considered were ease of collection, which represents whether or not variables are collectable, and frequency of occurrence, which represents whether or not changes made to input values constitute significant variables in terms of activity. The variables selected for providing personalized services were activity, weather, temperature, mean daily temperature, humidity, UV, fine dust, asthma and lung disease probability index, skin disease probability index, cadence, travel distance, mean heart rate, and sleep hours. The selected variables were classified according to the data characteristics. To predict activity, an LSTM RNN was built that uses the classified variables as input data and learns the dynamic characteristics of time series data. LSTM RNNs resolve the vanishing gradient problem that occurs in existing RNNs. They are classified into three different types according to data characteristics and constructed through connections among the LSTMs. The constructed neural network learns training data and predicts user activity. To evaluate the proposed model, the root mean square error (RMSE) was used in the performance evaluation of the user physical activity prediction method for which an autoregressive integrated moving average (ARIMA) model, a convolutional neural network (CNN), and an RNN were used. The results show that the proposed DC-LSTM RNN method yields an excellent mean RMSE value of 0.616. The proposed method is used for predicting significant activity considering the surrounding circumstances and user status utilizing the existing standardized activity prediction services. It can also be used to predict user physical activity and provide personalized healthcare based on the data collectable from mobile host devices.

• 대한토목학회논문집
• /
• 제40권3호
• /
• pp.303-314
• /
• 2020

수자원분야에서 이용되는 강우에 따른 유역의 수문학적 시스템, 도시지역 및 하천에 대한 수리학적 시스템은 비선형성이 강하고 많은 변수들을 포함하고 있다. 이러한 특성을 가진 시계열 자료에서 기계학습을 통한 예측은 예측시점 이전의 자료 특성을 반영하지 못하는 등 기본적인 신경망으로는 부족한 상황이 발생하기도 한다. 본 연구에서 적용할 강우-유출량과 같이 비선형성이 강하고 시간종속성이 높은 복잡한 시계열 자료를 예측하기 위해 신경망의 학습능력을 극대화한 순환형 동적 신경망(Recurrent Dynamic Neural Network)의 한 종류인 동시에, 시간 지연 신경망(Time-Delay Neural Network)의 특성을 가진 비선형 자기회귀(NARX, Nonlinear Autoregressive Exogenous Model) 인공신경망을 사용하였다. 이를 태화강 지방하천 구간에 적용하여 NARX 인공신경망의 시간 지연 매개변수를 10분에서 120분까지 조정하며 모의한 결과에 대해 여러 통계지표를 이용해 정량적으로 평가하였다. 그 결과 지연시간이 증가할수록 효율계수(NSE)가 0.530에서 0.988으로 증가하고, 평균제곱근편차(RMSE)가 379.9 ㎥/s에서 16.1 ㎥/s로 감소하는 등 정교한 예측이 가능함을 확인하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.712-715
• /
• 1997

In image jacobian based visual servoing, generally, inverse jacobian should be calculated by complicated coordinate transformations. These are required excessive computation and the singularity of the image jacobian should be considered. This paper presents a visual servoing to control the pose of the robotic manipulator for tracking and grasping 3-D moving object whose pose and motion parameters are unknown. Because the object is in motion tracking and grasping must be done on-line and the controller must have continuous learning ability. In order to estimate parameters of a moving object we use the kalman filter. And for tracking and grasping a moving object we use a fuzzy inference based reinforcement learning algorithm of dynamic recurrent neural networks. Computer simulation results are presented to demonstrate the performance of this visual servoing

• 한국생산제조학회지
• /
• 제19권3호
• /
• pp.394-401
• /
• 2010

In this article, we investigate a robust friction compensation scheme for the purpose of accomplishing precision positioning performance a servo mechanical system with nonlinear dynamic friction. To estimate the friction state and tackle robustness problem for uncertainty, a RFNN and reconstructed error compensator as well as a robust friction state observer are developed. The asymptotic stability of the series of friction compensation methodologies are verified from the Lyapunov's stability theory. Some simulations and experiments on a servo mechanical system were carried out to evaluate the effectiveness of the proposed control scheme.

• JSTS:Journal of Semiconductor Technology and Science
• /
• 제14권2호
• /
• pp.252-261
• /
• 2014

A Bayesian network (BN) based fault diagnosis framework for semiconductor etching equipment is presented. Suggested framework contains data preprocessing, data synchronization, time series modeling, and BN inference, and the established BNs show the cause and effect relationship in the equipment module level. Statistically significant state variable identification (SVID) data of etch equipment are preselected using principal component analysis (PCA) and derivative dynamic time warping (DDTW) is employed for data synchronization. Elman's recurrent neural networks (ERNNs) for individual SVID parameters are constructed, and the predicted errors of ERNNs are then used for assigning prior conditional probability in BN inference of the fault diagnosis. For the demonstration of the proposed methodology, 300 mm etch equipment model is reconstructed in subsystem levels, and several fault diagnosis scenarios are considered. BNs for the equipment fault diagnosis consists of three layers of nodes, such as root cause (RC), module (M), and data parameter (DP), and the constructed BN illustrates how the observed fault is related with possible root causes. Four out of five different types of fault scenarios are successfully diagnosed with the proposed inference methodology.

• International Journal of Computer Science & Network Security
• /
• 제23권8호
• /
• pp.177-189
• /
• 2023

Malware detection is an increasingly important operational focus in cyber security, particularly given the fast pace of such threats (e.g., new malware variants introduced every day). There has been great interest in exploring the use of machine learning techniques in automating and enhancing the effectiveness of malware detection and analysis. In this paper, we present a deep recurrent neural network solution as a stacked Long Short-Term Memory (LSTM) with a pre-training as a regularization method to avoid random network initialization. In our proposal, we use global and short dependencies of the inputs. With pre-training, we avoid random initialization and are able to improve the accuracy and robustness of malware threat hunting. The proposed method speeds up the convergence (in comparison to stacked LSTM) by reducing the length of malware OpCode or bytecode sequences. Hence, the complexity of our final method is reduced. This leads to better accuracy, higher Mattews Correlation Coefficients (MCC), and Area Under the Curve (AUC) in comparison to a standard LSTM with similar detection time. Our proposed method can be applied in real-time malware threat hunting, particularly for safety critical systems such as eHealth or Internet of Military of Things where poor convergence of the model could lead to catastrophic consequences. We evaluate the effectiveness of our proposed method on Windows, Ransomware, Internet of Things (IoT), and Android malware datasets using both static and dynamic analysis. For the IoT malware detection, we also present a comparative summary of the performance on an IoT-specific dataset of our proposed method and the standard stacked LSTM method. More specifically, of our proposed method achieves an accuracy of 99.1% in detecting IoT malware samples, with AUC of 0.985, and MCC of 0.95; thus, outperforming standard LSTM based methods in these key metrics.

• 한국공간구조학회논문집
• /
• 제20권2호
• /
• pp.51-58
• /
• 2020

Recently, deep learning that is the most popular and effective class of machine learning algorithms is widely applied to various industrial areas. A number of research on various topics about structural engineering was performed by using artificial neural networks, such as structural design optimization, vibration control and system identification etc. When nonlinear semi-active structural control devices are applied to building structure, a lot of computational effort is required to predict dynamic structural responses of finite element method (FEM) model for development of control algorithm. To solve this problem, an artificial neural network model was developed in this study. Among various deep learning algorithms, a recurrent neural network (RNN) was used to make the time history response prediction model. An RNN can retain state from one iteration to the next by using its own output as input for the next step. An eleven-story building structure with semi-active tuned mass damper (TMD) was used as an example structure. The semi-active TMD was composed of magnetorheological damper. Five historical earthquakes and five artificial ground motions were used as ground excitations for training of an RNN model. Another artificial ground motion that was not used for training was used for verification of the developed RNN model. Parametric studies on various hyper-parameters including number of hidden layers, sequence length, number of LSTM cells, etc. After appropriate training iteration of the RNN model with proper hyper-parameters, the RNN model for prediction of seismic responses of the building structure with semi-active TMD was developed. The developed RNN model can effectively provide very accurate seismic responses compared to the FEM model.

• 한국해양학회지:바다
• /
• 제28권4호
• /
• pp.133-142
• /
• 2023

이 논문에서는 RNN (Recurrent Neural Networks)-LSTM (Long Short-Term Memory) 을 적용하여 Lorenz 시스템을 예측하는 자료 기반 인공지능 모델을 구축하고, 이 모델이 미분방정식을 차분화하여 해를 구하는 역학 모델을 대체할 수 있는지 가능성을 진단하였다. 구축된 자료기반 모델이 초기 조건의 작은 교란이 근본적으로 다른 결과를 만들어내는 Lorenz 시스템의 카오스적인 특성을 반영한다는 것과, 시스템의 안정적인 두 개의 닻을 중심으로 운동하면서 전이 과정을 반복하는 특성, "결정론적 불규칙 흐름"의 특성, 분기 현상을 모사한다는 것을 확인하였다. 또한, 적분 시간 간격을 조절함으로써 전산자원을 절감할 수 있는 자료기반 모델의 장점을 보였다. 향후 자료기반 모델의 정교화와 자료기반 모델을 위한 자료동화 기법의 연구를 통해 자료기반 인공지능 모델의 활용성을 확대할 수 있을 것으로 기대한다.

• 한국지능시스템학회논문지
• /
• 제10권5호
• /
• pp.395-400
• /
• 2000

본 논문에서는 강화학습에 기반한 새로운 뉴로-퍼지 제어기를 제안한다. 시스템은 개체의 행동을 결정하는 뉴로-퍼지 제어기와 그 행동을 평가하는 동적 귀환 신경회로망으로 구성된다. 뉴로-퍼지 제어기의 후건부 소속함수는 강화학습을 한다. 한편, 유전자 알고리즘을 통하여 진화하는 동적 귀환 신경회로망은 환경으로부터 받는 외부 강화신호와 로봇의 상태로부터 내부강화 신호를 만들어낸다. 이 출력(내부강화신호)은 뉴로-퍼지 제어기의 교사신호로 사용되어 제어기가 학습을 지속하도록 만든다. 제안한 시스템은 미지의 환경에서 제어기의 최적화 및 적응에 사용할 수 있다. 제안한 알고리즘은 컴퓨터 시뮬레이션 상에서 자율 이동로봇의 장애물 회피에 적용하여 그 유효성을 확인한다.