• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제9권4호
• /
• pp.327-332
• /
• 2009

This paper proposes a method to improve the accuracy of a short-term electrical load forecasting (STLF) system based on neuro-fuzzy models. The proposed method compensates load forecasts based on the error obtained during the previous prediction. The basic idea behind this approach is that the error of the current prediction is highly correlated with that of the previous prediction. This simple compensation scheme using error information drastically improves the performance of the STLF based on neuro-fuzzy models. The viability of the proposed method is demonstrated through the simulation studies performed on the load data collected by Korea Electric Power Corporation (KEPCO) in 1996 and 1997.

• 한국가시화정보학회지
• /
• 제19권3호
• /
• pp.22-30
• /
• 2021

Ventilated cavity shapes by varying angle of attack of a circular cavitator were predicted based on Logvinovich's Independence Principle in order to verify the cavity shape prediction method. The prediction results were compared with model experiments conducted in the high-speed cavitation tunnel. In the prediction of the cavity centerline, the movement of the cavity centerline due to the effect of gravity and cavitator's angle of attack were well predicted. In the prediction of the cavity contour, it was found that the cavity edge prediction error increased as the angle of attack increased. The error of the upper cavity contour was small at the positive angle of attack, and the error of the lower cavity contour was small at the negative angle of attack.

• 대한전자공학회논문지TC
• /
• 제41권10호
• /
• pp.45-52
• /
• 2004

본 논문에서는 에러 예측회로를 사용하여 연집에러 입력시 성능저하를 보완한 경판정 비터비 디코더를 제안하였다. 비터비 디코더는 최대유사복호 알고리즘을 사용하므로 랜덤에러 입력시 정정능력이 뛰어나다. 반면에 연집에러 입력시 에러 정정능력이 매우 떨어지는 단점이 있다. 제안하는 에러 예측회로는 비터비 디코더의 연집에러에 대한 에러 정정특성을 향상시키는 기능으로 비터비 디코더에 에러가 입력됨에 따라 path metric값이 증가하는 것을 이용한다. Path metric의 최대값 증가량을 이용하여 연집에러 구간을 예측, 연집에러 구간에 대한 확률 값을 줄여준다. 제안된 알고리즘을 OFDM방식의 IEEE802.11a WLAN에 적용한 비터비 디코더는 AWGN채널에서는 기존의 비터비 디코더와 동일한 성능을 유지하며, 무선 채널 환경인 다중경로 페이딩 채널에서 발생할 수 있는 연집에러에 대하여 15% 개선된 성능을 보였다.

• Nuclear Engineering and Technology
• /
• 제54권4호
• /
• pp.1439-1448
• /
• 2022

Background: We investigated the feasibility of in vitro radiosensitivity prediction with gene expression using deep learning. Methods: A microarray gene expression of the National Cancer Institute-60 (NCI-60) panel was acquired from the Gene Expression Omnibus. The clonogenic surviving fractions at an absorbed dose of 2 Gy (SF2) from previous publications were used to measure in vitro radiosensitivity. The radiosensitivity prediction model was based on the convolutional neural network. The 6-fold cross-validation (CV) was applied to train and validate the model. Then, the leave-one-out cross-validation (LOOCV) was applied by using the large-errored samples as a validation set, to determine whether the error was from the high bias of the folded CV. The criteria for correct prediction were defined as an absolute error<0.01 or a relative error<10%. Results: Of the 174 triplicated samples of NCI-60, 171 samples were correctly predicted with the folded CV. Through an additional LOOCV, one more sample was correctly predicted, representing a prediction accuracy of 98.85% (172 out of 174 samples). The average relative error and absolute errors of 172 correctly predicted samples were 1.351±1.875% and 0.00596±0.00638, respectively. Conclusion: We demonstrated the feasibility of a deep learning-based in vitro radiosensitivity prediction using gene expression.

• 소음진동
• /
• 제8권6호
• /
• pp.1023-1029
• /
• 1998

When one attempts to construct a hologram. one finds that there are many sources of measurement errors. These errors are even amplified if one predicts the pressures close to the sources. The pressure estimation errors depend on the following parameters: the measurement spacing on the hologram plane. the prediction spacing on the prediction plane. and the distance between the hologram and the prediction plane. This raper analyzes quantitatively the errors when these are distributed irregularly on the hologram plane The sensor mismatch and inaccurate measurement location. position mismatch. are mainly addressed. In these cases. one can assume that the measurement is a sample of many measurement events. The bias and random error are derived theoretically. Then the relationship between the random error amplification ratio and the parameters mentioned above is examined quantitatively in terms of energy.

• 디지털산업정보학회논문지
• /
• 제5권3호
• /
• pp.141-145
• /
• 2009

In this paper, we propose an efficient lossless compression algorithm using spatial and temporal information. The proposed method obtains higher lossless compression of images than other lossless compression techniques. It is divided into two parts, a motion adaptation based predictor part and a residual error coding part. The proposed nonlinear predictor can reduce prediction error by learning from its past prediction errors. The predictor decides the proper selection of the spatial and temporal prediction values according to each past prediction error. The reduced error is coded by existing context coding method. Experimental results show that the proposed algorithm has better performance than those of existing context modeling methods.

• 한국소음진동공학회:학술대회논문집
• /
• 한국소음진동공학회 2007년도 추계학술대회논문집
• /
• pp.1325-1328
• /
• 2007

When we visualize the sound field radiated from a spherical sound source, spherical acoustic holography is proper among acoustic holography methods. However, there are measurement errors due to sensor position mismatch, sensor mismatch, directivity of sensor, and background noise. These errors are amplified if one predicts the pressures close to the sources: backward prediction. The goal of this paper is to quantitatively examine the effects of the error due to sensor position mismatch on acoustic pressure estimation. This paper deals with the cases of which the measurement deviations are distributed irregularly on the hologram plane. In such cases, one can assume that the measurement is a sample of many measurement events, and the cause of the measurement error is white noise on the hologram plane. Then the bias and random error are derived mathematically. In the results, it is found that the random error is important in the backward prediction. The relationship between the random error amplification ratio and the measurement parameters is derived quantitatively in terms of their energies.

• 응용통계연구
• /
• 제34권5호
• /
• pp.697-710
• /
• 2021

상대오차를 이용한 예측법은 상대오차(혹은 퍼센트오차)가 중요시되는 분야, 특히 계량경제학이나 소프트웨어 엔지니어링, 또는 정부기관 공식통계 부분에서 기존 예측방법 외에 선호되는 예측방법이다. 그 동안 상대오차를 이용한 예측법은 선형 혹은 비선형 회귀분석 뿐 아니라, 커널회귀를 이용한 비모수 회귀모형, 그리고 정상시계열분석에 이르기까지 그 범위가 확장되어 왔다. 그러나, 지금까지의 분석은 고정효과(fixed effect)만을 고려한 것이어서 임의효과(random effect)에 관한 상대오차 예측법에 대한 확장이 필요하였다. 본 논문의 목적은 상대오차예측법을 일반화선형혼합모형(GLMM)에 속한 감마회귀(gamma regression), 로그정규회귀(lognormal regression), 그리고 역가우스회귀(inverse gaussian regression)의 패널자료(panel data)에 적용시키는데 있다. 이를 위해 실제 자동차 보험회사의 손해액 자료를 사용하였고, 최량예측량과 최량상대오차예측량을 각각 적용-비교해 보았다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2001년도 하계종합학술대회 논문집(1)
• /
• pp.69-72
• /
• 2001

Blind equalization of transmission channel is important in communication areas and signal processing applications because it does not need training sequence, nor does it require a priori channel information. Recently, Tong et al. proposed solutions for this problem exploit the diversity induced by antenna array or time oversampling, leading to the second order statistics techniques, fur example, subspace method, prediction error method, and so on. The linear prediction error method is perhaps the most attractive in practice due to the insensitive to blind equalizer length mismatch as well as for its simple adaptive filter implementation. Unfortunately, the previous one-step prediction error method is known to be limited in arbitrary delay. In this paper, we induce the optimal delay, and propose the adaptive blind equalizer with multi-step linear prediction using RLS-type algorithm. Simulation results are presented to demonstrate the proposed algorithm and to compare it with existing algorithms.

• 한국IT서비스학회지
• /
• 제11권1호
• /
• pp.211-221
• /
• 2012

Many metric-based classification models have been proposed to predict fault-proneness of software module. This paper presents two prediction models using Bayesian classifier which is one of the most popular modern classification algorithms. Bayesian model based on Bayesian probability theory can be a promising technique for software quality prediction. This is due to the ability to represent uncertainty using probabilities and the ability to partly incorporate expert's knowledge into training data. The two models, Na$\ddot{i}$veBayes(NB) and Bayesian Belief Network(BBN), are constructed and dimensionality reduction of training data and test data are performed before model evaluation. Prediction accuracy of the model is evaluated using two prediction error measures, Type I error and Type II error, and compared with well-known prediction models, backpropagation neural network model and support vector machine model. The results show that the prediction performance of BBN model is slightly better than that of NB. For the data set with ambiguity, although the BBN model's prediction accuracy is not as good as the compared models, it achieves better performance than the compared models for the data set without ambiguity.