• Journal of the Korean Data and Information Science Society
• /
• 제7권1호
• /
• pp.13-19
• /
• 1996

In this paper, we study and compare two types of methods for classification when both continuous and categorical variables are used to describe each individual. One is neural network(NN) method using backpropagation learning(BPL). The other is logistic model(LM) method. Both the NN and LM are based on projections of the data in directions determined from interconnection weights.

• Journal of Electrical Engineering and Technology
• /
• 제4권3호
• /
• pp.389-395
• /
• 2009

Industrial motors are subject to incipient faults which, if undetected, can lead to motor failure. The necessity of incipient fault detection can be justified by safety and economical reasons. The technology of artificial neural networks has been successfully used to solve the motor incipient fault detection problem. This paper develops inexpensive, reliable, and noninvasive NN based incipient fault detection scheme for small and medium sized induction motors. Detailed design procedure for achieving the optimal NN model and Principal Component Analysis for dimensionality reduction is proposed. Overall thirteen statistical parameters are used as feature space to achieve the desired classification. GFFD NN model is designed and verified for optimal performance in fault identification on experimental data set of custom designed 2 HP, three phase 50 Hz induction motor.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 1996년도 춘계공동학술대회논문집; 공군사관학교, 청주; 26-27 Apr. 1996
• /
• pp.733-736
• /
• 1996

Neural Network(NN) is known to be suitable for forecasting corporate bankruptcy because of discriminant capability. Bankruptcy prediciton on NN by now has mostly been studied based on financial indices at specific point of time. However, the financial profile of corporates fluctuates within a certain range with the elapse of time. Besides, we need a lot of data of different bankrupt types in order to apply NN for better bankruptcy prediciton. Therefore, we have decided to focus on textiles and clothing industries for bankruptcy prediction with limited data. One part of the collected data was used for training and calibration, and the other was used for verification. The model makes a learning with extended data from financial indices at specific point of time. The trained model has been tested and we could get a high hitting ratio relatively.

• 경영과학
• /
• 제14권2호
• /
• pp.91-111
• /
• 1997

Neural Network(NN) is known to be suitable for forecasting corporate bankruptcy because of discriminant capability. Bandkruptcy prediction on NN by now has mostly been studied based on financial indices at specific point of time. However, the financial profile of corporates fluctuates within a certain range with the elapse of time. Besides, we need a lot of data of different bankrupt types in order to apply NN for better bankruptcy prediction. Therefore, We have decided to focus on textile and clothing industries for bankruptcy prediction with limited data. One part of the collected data was used for training and calibration, and the other was used for verification. The model makes a learning with extended data from financial indices at specific point of time. The trained model has been tested and we could get a high hitting ratio relatively.

• 한국경영과학회지
• /
• 제14권2호
• /
• pp.91-91
• /
• 1989

Neural Network(NN) is known to be suitable for forecasting corporate bankruptcy because of discriminant capability. Bandkruptcy prediction on NN by now has mostly been studied based on financial indices at specific point of time. However, the financial profile of corporates fluctuates within a certain range with the elapse of time. Besides, we need a lot of data of different bankrupt types in order to apply NN for better bankruptcy prediction. Therefore, We have decided to focus on textile and clothing industries for bankruptcy prediction with limited data. One part of the collected data was used for training and calibration, and the other was used for verification. The model makes a learning with extended data from financial indices at specific point of time. The trained model has been tested and we could get a high hitting ratio relatively.

• 전자공학회논문지B
• /
• 제33B권9호
• /
• pp.97-106
• /
• 1996

In this paper, we propose a face recognition algorithm using the hidden markov model and neural networks (HMM-NN). In the preprocessing stage, we find edges of a face using the locally adaptive threshold (LAT) scheme and extract features based on generic knowledge of a face, then construct a database with extracted features. In the training stage, we generate HMM parameters for each person by using the forward-backward algorithm. In the recognition stage, we apply probability vlaues calculated by the HMM to subsequent neural networks (NN) as input data. Computer simulation shows that the proposed HMM-NN algorithm gives higher recognition rate compared with conventional face recognition algorithms.

• 정보처리학회논문지D
• /
• 제8D권3호
• /
• pp.241-246
• /
• 2001

소프트웨어공학에서 소프트웨어 측정분야는 30년 이상 수많은 연구가 있어 왔으나 아직까지 구체적인 소프트웨어 비용추정 모델이 없는 실정이다. 만약 소프트웨어 비용-개발노력을 측정하려면 소프트웨어 규모를 추정해야 한다. 많은 소프트웨어 척도가 개발되었지만 가장 일반적인 척도가 LOC(line of code)와 FPA(Function Point Analysis)이다. FPA는 소프트웨어 규모를 측정하는데 LOC를 사용할 때의 단점을 극복할 수 있는 기법이다. 본 논문은 FP와 기능 구성요소 형태들로 측정된 소프트웨어 규모로 소프트웨어 개발 노력을 추정하는 신경망 모델을 제안한다. 24개 소프트웨어 개발 프로젝트 사례연구를 통해 적합한 신경망 모델을 제시하였다. 또한, 희귀분석 모델과 신경망 모델을 비교하여 신경망 모델의 추정 정확성이 보다 좋음을 보였다.

• Smart Structures and Systems
• /
• 제16권6호
• /
• pp.1133-1145
• /
• 2015

A new neural network (NN) predictive controller (NNPC) algorithm has been developed and tested in the computer simulation of active control of a nonlinear structure. In the present method an NN is used as a predictor. This NN has been trained to predict the future response of the structure to determine the control forces. These control forces are calculated by minimizing the difference between the predicted and desired responses via a numerical minimization algorithm. Since the NNPC is very time consuming and not suitable for real-time control, it is then used to train an NN controller. To consider the effectiveness of the controller on probability of damage, fragility curves are generated. The approach is validated by using simulated response of a 3 story nonlinear benchmark building excited by several historical earthquake records. The simulation results are then compared with a linear quadratic Gaussian (LQG) active controller. The results indicate that the proposed algorithm is completely effective in relative displacement reduction.

• 한국지능시스템학회논문지
• /
• 제6권3호
• /
• pp.40-50
• /
• 1996

By using the neural networks(NN) as system identifier, the on-line self tuning method for fuzzy controller(FC) is proposed. In theis method, the learning of NN is carried out during control operation of FC and the cinsequent parameters of FC is tuned on-line automatically by means of system output errors backpropagated through NN. The Sugeno fuzzy model with constants as consequent parameters is selected for simplifying computation. In procedures of parameter tuning, the gradient descent method is used and the gradient vectors for adjusting the weight of NN are transferred as controller output errors. To evaluate the performance, the proposed method is applied to the inverted pendulum system.

• 한국멀티미디어학회논문지
• /
• 제25권6호
• /
• pp.783-792
• /
• 2022

Versatile Video Coding (VVC) is the latest video coding standard, which had been developed by the Joint Video Experts Team (JVET) of ITU-T Video Coding Experts Group (VCEG) and ISO/IEC Moving Picture Experts Group (MPEG) in 2020. Although VVC can provide powerful coding performance, it requires tremendous computational complexity to determine the optimal block structures during the encoding process. In this paper, we propose a fast ternary tree decision method using two neural networks with 7 nodes as input vector based on the multi-layer perceptron structure, names STH-NN and STV-NN. As a training result of neural network, the STH-NN and STV-NN achieved accuracies of 85% and 91%, respectively. Experimental results show that the proposed method reduces the encoding complexity up to 25% with unnoticeable coding loss compared to the VVC test model (VTM).