• Journal of the Korean Data and Information Science Society
• /
• v.18 no.4
• /
• pp.915-925
• /
• 2007

Multivariate GARCH has been useful to model dynamic relationships between volatilities arising from each component series of multivariate time series. Methodologies including EWMA(Exponentially weighted moving-average model), DVEC(Diagonal VEC model), BEKK and CCC(Constant conditional correlation model) models are comparatively reviewed for bivariate time series. In addition, these models are applied to evaluate VaR(Value at Risk) and to construct joint prediction region. To illustrate, bivariate stock prices data consisting of Samsung Electronics and LG Electronics are analysed.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.5
• /
• pp.555-561
• /
• 2009

The established mathematical modeling methods have limitation to know the hydraulic characteristics at the wastewater treatment plant which are complex and nonlinear systems. So, an artificial neural network (ANN) model based on hydraulic characteristics is applied for modeling wastewater quality of a full-scale wastewater treatment plant using DNR (Daewoo nutrient removal) process. ANN was trained using data which are influents (TSS, BOD, COD, TN, TP) and effluents (COD, TN, TP) components in a year, and predicted the effluent results based on the training. To raise the efficiency of prediction, inputs of ANN are added the influent and effluent information that are in yesterday and the day before yesterday. The results of training data tend to have high accuracy between real value and predicted value, but test data tend to have lower accuracy. However, the more hydraulic characteristics are considered, the results become more accuracy.

• KIPS Transactions on Software and Data Engineering
• /
• v.6 no.9
• /
• pp.437-444
• /
• 2017

Although many studies tried to predict movie revenues in the last decade, the main focus is still to learn an efficient forecast model to fit the box-office revenues. However, the previous works lack the analysis about why the prediction errors occur, and no method is proposed to reduce the errors. In this paper, we consider the prediction error comes from the competition between the movies that are released in the same period. Our purpose is to analyze the competition value for a movie and to predict how much it will be affected by other competitors so as to improve the performance of movie box-office prediction. In order to predict the competition value, firstly, we classify its sign (positive/negative) and compute the probability of positive sign and the probability of negative sign. Secondly, we forecast the competition value by regression under the condition that its sign is positive and negative respectively. And finally, we calculate the expectation of competition value based on the probabilities and values. With the predicted competition, we can adjust the primal predicted box-office. Our experimental results show that predictive competition can help improve the performance of the forecast.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2008.03a
• /
• pp.1365-1371
• /
• 2008

Based on the data obtained from field investigation and soil testing to slope hazards occurrence section and non-occurrence section in gneiss area, a prediction technique was developed by the use of a decision tree model. The slope hazards data of Seoul and Kyonggi Province were 104 sections in gneiss area. The number of data applied in developing prediction model was 61 sections except a vacant value. The statistical analyses using the decision tree model were applied to the entrophy index. As the results of analyses, a slope angle, a degree of saturation and an elevation were selected as the classification standard. The prediction model of decision tree using entrophy index is most likely accurate. The classification standard of the selected prediction model is composed of the slope angle, the degree of saturation and the elevation from the first choice stage. The classification standard values of the slope angle, the degree of saturation and elevation are $17.9^{\circ}$, 52.1% and 320m, respectively.

• The Journal of Bigdata
• /
• v.5 no.2
• /
• pp.215-229
• /
• 2020

The interest in machine learning is growing in all industries, but it is difficult to apply it to real-world tasks because of inexplicability. This paper introduces a case of developing a financial customer churn prediction model for a securities company, and introduces the research results on an attempt to develop a machine learning model that can be explained using the SHAP Value methodology and derivation of interpretability. In this study, a total of six customer churn models are compared and analyzed, and the cause of customer churn is inferred through the classification and data analysis of SHAP Value and the type of customer asset change. Based on the results of this study, it would be possible to use it as a basis for comprehensive judgment, such as using the Value of the deviation prediction result that can infer the cause of the marketing manager's actual customer marketing in the future and establishing a target marketing strategy for each customer.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.11 no.6
• /
• pp.40-48
• /
• 2012

In this paper, we suggests methods for determining optimal representative value and the optimal size of historical data for reliable travel speed prediction. To evaluate the performance of the proposed method in real world environments, we did field tests at four roadway links in Seoul on Tuesday and Sunday. According to the results of applying the methods to historical data of Central Traffic Information Center, the optimal representative value were analyzed to be average and weighted average. Second, it was analyzed that 2 months data is the optimal size of historical data used for travel speed prediction.

• International journal of advanced smart convergence
• /
• v.11 no.2
• /
• pp.30-37
• /
• 2022

Because fine dust (PM10) has a close effect on the environment, fine dust generated in the climate and living environment has a bad effect on the human body. In this study, the LSTM model was applied to predict and analyze the effect of fine dust on Gwangju Metropolitan City in Korea. This paper uses prediction values of input variables selected through correlation analysis to confirm fine dust prediction performance. In this paper, data from the Gwangju Metropolitan City area were collected to measure fine dust. The collection period is one year's worth of data was used from january to December of 2021, and the test data was conducted using three-month data from January to March of 2022. As a result of this study, in the as a result of predicting fine dust (PH10) and ultrafine dust (PH2.5) using the LSTM model, the RMSE was 4.61 and the test result value was as low as 4.37. This reason is judged to be the result of the contents of the one-year sample.

• Proceedings of the IEEK Conference
• /
• 2003.11b
• /
• pp.235-238
• /
• 2003

Conventional stride predictor is useful for predicting data values which vary by a constant value. However, when the data values of shift, multiplication, and division instructions are predicted, the stride predictor can't show the best performance. Thus, we propose predictor using stride and shift to improve predictability. The predictor using stride and shift takes advantage of shift values as well as stride values, so that the overall coverage of prediction increases.

• Journal of Korea Trade
• /
• v.25 no.4
• /
• pp.17-36
• /
• 2021

Purpose - This study provides useful information to stakeholders by forecasting the tramp shipping market, which is a completely competitive market and has a huge fluctuation in freight rates due to low barriers to entry. Moreover, this study provides the most effective parameters for Baltic Dry Index (BDI) prediction and an optimal model by analyzing and comparing deep learning models such as the artificial neural network (ANN), recurrent neural network (RNN), and long short-term memory (LSTM). Design/methodology - This study uses various data models based on big data. The deep learning models considered are specialized for time series models. This study includes three perspectives to verify useful models in time series data by comparing prediction accuracy according to the selection of external variables and comparison between models. Findings - The BDI research reflecting the latest trends since 2015, using weekly data from 1995 to 2019 (25 years), is employed in this study. Additionally, we tried finding the best combination of BDI forecasts through the input of external factors such as supply, demand, raw materials, and economic aspects. Moreover, the combination of various unpredictable external variables and the fundamentals of supply and demand have sought to increase BDI prediction accuracy. Originality/value - Unlike previous studies, BDI forecasts reflect the latest stabilizing trends since 2015. Additionally, we look at the variation of the model's predictive accuracy according to the input of statistically validated variables. Moreover, we want to find the optimal model that minimizes the error value according to the parameter adjustment in the ANN model. Thus, this study helps future shipping stakeholders make decisions through BDI forecasts.

• Journal of Korea Multimedia Society
• /
• v.24 no.9
• /
• pp.1251-1260
• /
• 2021

This paper discusses missing data processing using simple moving average (SMA) and kalman filter. Also SMA and kalman predictive value are made a comparative study. Time series analysis is a generally method to deals with time series data in photovoltaic field. Photovoltaic system records data irregularly whenever the power value changes. Irregularly recorded data must be transferred into a consistent format to get accurate results. Missing data results from the process having same intervals. For the reason, it was imputed using SMA and kalman filter. The kalman filter has better performance to observed data than SMA. SMA graph is stepped line graph and kalman filter graph is a smoothing line graph. MAPE of SMA prediction is 0.00737%, MAPE of kalman prediction is 0.00078%. But time complexity of SMA is O(N) and time complexity of kalman filter is O(D²) about D-dimensional object. Accordingly we suggest that you pick the best way considering computational power.