• 전기학회논문지
• /
• 제67권1호
• /
• pp.114-123
• /
• 2018

This paper concerns a design and learning method of softmax function neural networks based on K-means clustering. The partial discharge data Information is preliminarily processed through simulation using an Epoxy Mica Coupling sensor and an internal Phase Resolved Partial Discharge Analysis algorithm. The obtained information is processed according to the characteristics of the pattern using a Motor Insulation Monitoring System program. At this time, the processed data are total 4 types that void discharge, corona discharge, surface discharge and slot discharge. The partial discharge data with high dimensional input variables are secondarily processed by principal component analysis method and reduced with keeping the characteristics of pattern as low dimensional input variables. And therefore, the pattern classifier processing speed exhibits improved effects. In addition, in the process of extracting the partial discharge data through the MIMS program, the magnitude of amplitude is divided into the maximum value and the average value, and two pattern characteristics are set and compared and analyzed. In the first half of the proposed partial discharge pattern classifier, the input and hidden layers are classified by using the K-means clustering method and the output of the hidden layer is obtained. In the latter part, the cross entropy error function is used for parameter learning between the hidden layer and the output layer. The final output layer is output as a normalized probability value between 0 and 1 using the softmax function. The advantage of using the softmax function is that it allows access and application of multiple class problems and stochastic interpretation. First of all, there is an advantage that one output value affects the remaining output value and its accompanying learning is accelerated. Also, to solve the overfitting problem, L2-normalization is applied. To prove the superiority of the proposed pattern classifier, we compare and analyze the classification rate with conventional radial basis function neural networks.

• 한국정보통신학회논문지
• /
• 제3권2호
• /
• pp.313-319
• /
• 1999

4층구조의 다층퍼셉트론을 변형하여 3 종류의 다층회귀예측신경망을 구성하고, 예측차수, 두 은닉층의 뉴런개수, 연결세기의 초기치 및 전달함수 변화에 따른 각 망의 음성인식성능을 실험을 통해 각각 비교 분석한다. 실험결과에 의하면, 다층회귀신경망이 다층퍼셉트론에 비해 음성인식성능이 우수하다. 그리고 구조적으로는 상위은닉층의 출력을 하위은닉층으로 회귀할 때 인식성능이 가장 우수하며, 각 망 공히 상, 하위은닉층의 뉴런 10 혹은 15개, 예측차수 3 혹은 4차일 때 인식률이 양호하다. 학습시 연결세기의 초기치를 -0.5에서 0.5사이로 설정하고, 하위은닉층에서 단극성 시그모이드 전달함수를 사용할 때 인식성능이 더욱 향상된다.

• 전기학회논문지
• /
• 제56권7호
• /
• pp.1324-1330
• /
• 2007

In this paper, we propose pattern recognition algorithm for activities of daily living by adopting extreme learning machine based on single layer feedforward networks(SLFNs) to the signal from bidirectional accelerometer. For activity classification, 20 persons are participated and we acquire 6, types of signals at standing, walking, running, sitting, lying, and falling. Then, we design input vector using reduced model for ELM input. In ELM classification results, we can find accuracy change by increasing the number of hidden neurons. As a result, we find the accuracy is increased by increasing the number of hidden neuron. ELM is able to classify more than 80 % accuracy for experimental data set when the number of hidden is more than 20.

• Journal of Positioning, Navigation, and Timing
• /
• 제8권4호
• /
• pp.193-200
• /
• 2019

The auto-encoder network which is a good candidate to handle the modeling of the signal strength attenuation is designed for denoising and compensating the distortion of the received data. It provides a non-linear mapping function by iteratively learning the encoder and the decoder. The encoder is the non-linear mapping function, and the decoder demands accurate data reconstruction from the representation generated by the encoder. In addition, the adaptive network width which supports the automatic generation of new hidden nodes and pruning of inconsequential nodes is also implemented in the proposed algorithm for increasing the efficiency of the algorithm. Simulation results show that the proposed method can improve the neural network training surface to achieve the highest possible accuracy of the signal modeling compared with the conventional modeling method.

• 대한교통학회지
• /
• 제13권1호
• /
• pp.95-110
• /
• 1995

n this research neural -based model was developed to forecast link travel times , And it is also compared wiht other time series forecasting models such as Box-Jenkins model, Kalman filter model. These models are validated to evaluate the accuracy of models with real time series data gathered by the license plate method. Neural network's convergency and generalization were investigated by modifying learning rate, momentum term and the number of hidden layer units. Through this experiment, the optimum configuration of the nerual network architecture was determined. Optimumlearining rate, momentum term and the number of hidden layer units hsow 0.3, 0.5, 13 respectively. It may be applied to DRGS(dynamic route guidance system) with a minor modification. The methods are suggested at the condlusion of this paper, And there is no doubt that this neural -based model can be applied to many other itme series forecating problem such as populationforecasting vehicel volume forecasting et .

• 대한전기학회논문지
• /
• 제43권6호
• /
• pp.1027-1035
• /
• 1994

This paper aims to make an artificial arm control strategy. For this, we propose a new feature extraction method and design artificial neural network for the functional classification of myoelectric signal(MES). We first transform the two channel myoelectric signals (MES) for biceps and triceps into frequency domain using fast Fourier transform (FFT). And features were obtained by comparing the magnitudes of ensemble spectrum data and used as inputs to the three-layer neural network for the learning. By changing the number of units in hidden layer of neural network we observed the improvement of classification performance. To observe the effeciency of the proposed scheme we performed experiments for classification of six arm functions to the three subjects. And we obtained on average 94[%] the ratio of classification.

• 한국멀티미디어학회논문지
• /
• 제24권6호
• /
• pp.796-804
• /
• 2021

In computer vision, visual tracking method addresses the problem of localizing an specific object in video sequence according to the bounding box. In this paper, we propose a tracking method by introducing the feature correlation comparison into the siamese network to increase its matching identification. We propose a way to compute location of object to improve matching performance by a correlation operation, which locates parts for solving the searching problem. The higher layer in the network can extract a lot of object information. The lower layer has many location information. To reduce error rate of the object center point, we built a siamese network that extracts the distribution and location information of target objects. As a result of the experiment, the average center error rate was less than 25%.

• 정보교육학회논문지
• /
• 제24권2호
• /
• pp.157-165
• /
• 2020

소프트웨어교육에 대한 관심이 높아지면서 소프트웨어교육의 교수·학습 방법 및 평가에 대한 논의도 같이 활발해지고 있다. 현재 이루어지고 있는 소프트웨어교육 수업 방법의 문제는 교수자가 학습자의 컴퓨터에서 진행되고 있는 코딩의 내용을 실시간으로 파악할 수 없다는 것이다. 이에 따라 교수자는 적시에 학습자에게 피드백을 주는데 한계가 있다. 이 문제를 극복하기 위하여 본 연구에서는 학습분석 기법을 적용하여 엔트리 기반의 실시간 학습자 코딩 상황을 파악하고 교수자에게 전달하는 소프트웨어교육 지원 시스템을 개발하고, 학습중에 수집되는 데이터를 Hadoop 시스템을 통하여 시각화는 체제를 구현하였다. 소프트웨어교육 지원 시스템은 교사와 학습자가 접속하는 표현 계층과 코드를 분석하고 구조화하여 평가하는 비즈니스 계층, 그리고 학급정보, 계정 정보, 학습정보 등을 저장하는 DB 계층을 포함하고 있다. 교수자는 미리 학습할 내용을 소프트웨어교육 지원 시스템에 설정하는 것이 가능하고, 저장된 코드와 학생들의 코드를 비교한 데이터를 기반으로 하여 컴퓨팅 사고력 요소 루브릭을 통해 학습자의 성취율을 비교·분석할 수 있다.

• 문화기술의 융합
• /
• 제9권4호
• /
• pp.487-494
• /
• 2023

본 연구에서는 무료 딥러닝 도구인 R과 텐서플로우에 대한 성능 비교를 수행하였다. 실험에서는 각 도구를 사용하여 6종류의 심층 신경망을 구축하고 10년간의 한국 온도 데이터셋을 사용하여 신경망을 학습시켰다. 구축된 신경망의 입력층 노드 갯수는 10개, 출력층은 5개로 설정 하였으며, 은닉층은 5, 10, 20개로 설정하여 실험을 진행 하였다. 학습 데이터는 2013년 3월 1일부터 2023년 3월 29일까지 서울시 강남구에서 수집된 온도 데이터 3681건을 사용하였다. 성능 비교를 위해, 학습된 신경망을 사용하여, 5일간의 온도를 예측하고 예측된 값과 실제값을 사용하여 평균 제곱근 오차(root mean square error, RMSE)값을 측정하였다. 실험결과, 은닉층이 1개인 경우, R의 학습 오차는 0.04731176이었으며, 텐서플로우는 0.06677193으로 측정되었으며, 은닉층이 2개인 경우에는 R이 0.04782134, 텐서플로 우는 0.05799060로 측정되었다. 전체적으로 R이 더 우수한 성능을 보였다. 우리는 기계학습을 처음 접하는 사용자들에게 두 도구에 대한 정량적 성능 정보를 제공함으로써, 도구 선택에서 발생하는 어려움을 해소하고자 하였다.

• 한국전산구조공학회:학술대회논문집
• /
• 한국전산구조공학회 1998년도 봄 학술발표회 논문집
• /
• pp.216-225
• /
• 1998

Although artificial neural network based on backpropagation algorithm is an excellent system simulator, it has still unsolved problems of its structure-decision and learning method. That is, we cannot find a general approach to decide the structure of the neural network and cannot train it satisfactorily because of the local optimum point which it frequently falls into. In addition, although there are many successful applications using backpropagation learning algorithm, there are few efforts to improve the learning algorithm itself. In this study, we suggest a general way to construct the hidden layer of the neural network using binary genetic algorithm and also propose the various learning methods by which the global minimum value of the teaming error can be obtained. A XOR problem and line heating problems are investigated as examples.