• Journal of the Korea Safety Management & Science
• /
• v.18 no.4
• /
• pp.161-169
• /
• 2016

In this research, we propose efficient demand forecasting scheme for intermittent demand. For this purpose, we first extensively analyze the drawbacks of the existing forecasting methods such as Croston method and Syntetos-Boylan approximation, then using these findings we propose the new demand forecasting method. Our goal is to develop forecasting method robust across many situations, not necessarily optimal for a limited number of specific situations. For this end, we adopt combining forecasting method that utilizes unbiased forecasting methods such as simple exponential smoothing and simple moving average. Various simulation results show that the proposed forecasting method performed better than the existing forecasting methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.78-81
• /
• 2021

Through the smart factory construction project, sensors can be installed in manufacturing production facilities and various process data can be collected in real time. Through this, research on real-time facility anomaly detection is being actively conducted to reduce production interruption due to facility abnormality in the manufacturing process. In this paper, to detect abnormalities in production facilities, the manufacturing data was applied to deep learning models Autoencoder(AE), VAE(Variational Autoencoder), and AAE(Adversarial Autoencoder) to derive the results. Manufacturing data was used as input data through a simple moving average technique and preprocessing process, and performance analysis was conducted according to the window size of the simple movement average technique and the feature vector size of the AE model.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.377-396
• /
• 2023

As the cryptocurrency market continues to grow, it has developed into a new financial market. The need for investment strategy research on the cryptocurrency market is also emerging. This study aims to conduct an empirical analysis on an investment methodology of cryptocurrency that combines short-term trading strategy and deep learning. Daily price data of the Ethereum was collected through the API of Upbit, the Korean cryptocurrency exchange. The investment performance of the experimental model was analyzed by finding the optimal parameters based on past data. The experimental model is a volatility breakout strategy(VBS), a Long Short Term Memory(LSTM) model, moving average cross strategy and a combined model. VBS is a short-term trading strategy that buys when volatility rises significantly on a daily basis and sells at the closing price of the day. LSTM is suitable for time series data among deep learning models, and the predicted closing price obtained through the prediction model was applied to the simple trading rule. The moving average cross strategy determines whether to buy or sell when the moving average crosses. The combined model is a trading rule made by using derived variables of the VBS and LSTM model using AND/OR for the buy conditions. The result shows that combined model is better investment performance than the single model. This study has academic significance in that it goes beyond simple deep learning-based cryptocurrency price prediction and improves investment performance by combining deep learning and short-term trading strategies, and has practical significance in that it shows the applicability in actual investment.

• Journal of the Korean Data and Information Science Society
• /
• v.20 no.2
• /
• pp.251-260
• /
• 2009

In this study, we propose a new estimation method based on autocovariance for selecting optimal estimators of the regression coefficients in the simple linear regression model. Although this method does not seem to be intuitively attractive, these estimators are unbiased for the corresponding regression coefficients. When the exploratory variable takes the equally spaced values between 0 and 1, under mild conditions which are satisfied when errors follow an autoregressive moving average model, we show that these estimators have asymptotically the same distributions as the least squares estimators. Additionally, under the same conditions as before, we provide a self-contained proof that these estimators converge in probability to the corresponding regression coefficients.

• Progress in Medical Physics
• /
• v.28 no.4
• /
• pp.181-189
• /
• 2017

The conventional delivery quality assurance (DQA) process for RapidArc (Varian Medical Systems, Palo Alto, USA), has the limitation that it measures and analyzes the dose in a phantom material and cannot analyze the dosimetric changes under the motional organ condition. In this study, a DQA method was designed to overcome the limitations of the conventional DQA process for internal target volume (ITV) based RapidArc. The dynamic DQA measurement device was designed with a moving phantom that can simulate variable target motions. The dose distribution in the real volume of the target and organ-at-risk (OAR)s were reconstructed using 3DVH with the ArcCHECK (SunNuclear, Melbourne, USA) measurement data under the dynamic condition. A total of 10 ITV-based RapidArc plans for liver-cancer patients were analyzed with the designed dynamic DQA process. The average pass rate of gamma evaluation was $81.55{\pm}9.48%$ when the DQA dose was measured in the respiratory moving condition of the patient. Appropriate method was applied to correct the effect of moving phantom structures in the dose calculation, and DVH data of the real volume of target and OARs were created with the recalculated dose by the 3DVH program. We confirmed the valid dose coverage of a real target volume in the ITV-based RapidArc. The variable difference of the DVH of the OARs showed that dose variation can occur differently according to the location, shape, size and motion range of the target. The DQA process devised in this study can effectively evaluate the DVH of the real volume of the target and OARs in a respiratory moving condition in addition to the simple verification of the accuracy of the treatment machine. This can be helpful to predict the prognosis of treatment by the accurate dose analysis in the real target and OARs.

• Convergence Security Journal
• /
• v.11 no.3
• /
• pp.19-24
• /
• 2011

Software failure time presented in the literature exhibit either constant monotonic increasing or monotonic decreasing, For data analysis of software reliability model, data scale tools of trend analysis are developed. The methods of trend analysis are arithmetic mean test and Laplace trend test. Trend analysis only offer information of outline content. In this paper, we discuss forecasting failure time case of failure time censoring. In this study, time series analys is used in the simple moving average and weighted moving averages, exponential smoothing method for predict the future failure times, Empirical analysis used interval failure time for the prediction of this model. Model selection using the mean square error was presented for effective comparison.

• Journal of the Korean Operations Research and Management Science Society
• /
• v.38 no.2
• /
• pp.75-93
• /
• 2013

In applying Markowitz's portfolio selection model to the stock market, we developed a comprehensive investment decision-making framework including key inputs for portfolio theory (i.e., individual stocks' expected rate of return and covariance) and minimum required expected return. For estimating the key inputs of our decision-making framework, we utilized an exponentially weighted moving average (EWMA) which places more emphasis on recent data than the conventional simple moving average (SMA). We empirically analyzed the investment results of the decision-making framework with the same 15 stocks in Samsung Group Funds found in the Korean stock market between 2007 and 2011. This five-year investment horizon is marked by global financial crises including the U.S. subprime mortgage crisis, the collapse of Lehman Brothers, and the European sovereign-debt crisis. We measure portfolio performance in terms of rate of return, standard deviation of returns, and Sharpe ratio. Results are compared with the following benchmarks : 1) KOSPI, 2) Samsung Group Funds, 3) Talmudic portfolio based on the na$\ddot{i}$ve 1/N rule, and 4) Markowitz's model with SMA. We performed sensitivity analyses on all the input parameters that are necessary for designing an investment decision-making framework : smoothing constant for EWMA, minimum required expected return for the portfolio, and portfolio rebalancing period. In conclusion, appropriate use of the comprehensive investment decision-making framework based on the Markowitz's model integrated with EWMA proves to achieve outstanding performance compared to the benchmarks.

• Journal of Korea Multimedia Society
• /
• v.10 no.11
• /
• pp.1530-1537
• /
• 2007

In this paper, we implemented a speech emotion recognition system that can distinguish human emotional states from recorded speech captured by a single microphone and classify them into four categories: neutrality, happiness, sadness and anger. In general, a speech recorded with a microphone contains background noises due to the speaker environment and the microphone characteristic, which can result in serious system performance degradation. In order to minimize the effect of these noises and to improve the system performance, a MA(Moving Average) filter with a relatively simple structure and low computational complexity was adopted. Then a SFS(Sequential Forward Selection) feature optimization method was implemented to further improve and stabilize the system performance. For speech emotion classification, a SVM pattern classifier is used. The experimental results indicate the emotional classification performance around 65% in the computer simulation and 62% on the ARM platform.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.8A
• /
• pp.671-678
• /
• 2005

This paper proposed symbol rate method which does not require a priori knowledge on the symbol rate and simplified modulation identification method to classify BPSK, QPSK, 8PSK signal. In order to estimate the unknown symbol rate, sliding FFT and simple moving average to estimate the spectrum of the signals is utilized, and sliding window and decimation, LPF blcok to estimate the proper symbol rate is used. Although conventional modulation ID method must use SNR value as the test statistics, the receiver cannot estimate the SNR value since the receiver cannot know the modulation type at the start of communication, and bit resolution is high due to using nonlinear function such as log, cosh. Therefore, we proposed the simplified fixed SNR value method. The performance of symbol rate estimation and modulation ID is shown using Monte Carlo computer simulation. This paper show that symbol rate estimation also has good performance in low SNR, and proposed simplified fixed SNR method has almost equivalent performance compared to conventional method.

• Journal of Korean Institute of Industrial Engineers
• /
• v.29 no.2
• /
• pp.135-144
• /
• 2003

Traffic performance in a microcellular system is much more affected by cell dwell time and channel holding time in each cell. Cell dwell time of a call is characterized by its mobility pattern, i.e., stochastic changes of moving speed and direction. Cell dwell time provides important information for other analyses on traffic performance such as channel holding time, handover rate, and the average number of handovers per call. In the next generation mobile communication system, the cell size is expected to be much smaller than that of current one to accommodate the increase of user demand and to achieve high bandwidth utilization. As the cell size gets small, traffic performance is much more affected by variable mobility of users, especially by that of pedestrians. In previous work, analytical models are based on simple probability models. They provide sufficient accuracy in a simple second-generation cellular system. However, the role of them is becoming invalid in a picocellular environment where there are rapid change of network traffic conditions and highly random mobility of pedestrians. Unlike in previous work, we propose an improved probability model evolved from so-called Random walk model in order to mathematically formulate variable mobility of pedestrians and analyze the traffic performance. With our model, we can figure out variable characteristics of pedestrian mobility with stochastic correlation. The above-mentioned traffic performance measures are analyzed using our model.