• The Research Journal of the Costume Culture
• /
• v.18 no.6
• /
• pp.1274-1290
• /
• 2010

This study analyzed the design elements and fashion images of women's knitwear in collections of Paris, Milan, London and New York between 2003 and 2008, and examined knitwear trends in an effort to verify whether knitwear trends are repeated in certain cycles, whether they show complicated patterns in cycles and yet occur in quasi cycles, or whether they occur non-periodically in complicated forms of chaotic cycles. Trend cycle analysis results are deemed to identify the time series attribute of knit fashions. It also sought to categorize the attribute of various factors influencing knitwear trends with a view to determining relevancy between design elements, and to present the direction of predicting knitwear fashion trends and the progression of short-term knitwear trends. This study reached the following conclusion. According to design elements or fashion images, knitwear fashion trends occur in cycles, quasi cycles, non-periodical cycles. These cyclic characteristics can be used as scientific data for planning knitwear products. The study confirmed close relevancy between fashion images and fashion elements. It identified close relevancy between designs with similar fashion elements and images through coordinates by year and season, and it is possible to make short-term prediction of trend direction through the flow of coordinates. Time series data were insufficient, thereby making it difficult to perfectly verify chaos indices and giving limitations to this study. A study with more time series data will produce a more effective method of predicting and using knitwear fashion trends.

• Korean Journal of Remote Sensing
• /
• v.36 no.4
• /
• pp.515-525
• /
• 2020

This study investigates the potential of bidirectional long short-term memory (Bi-LSTM) for efficient modeling of temporal information in crop classification using multitemporal remote sensing images. Unlike unidirectional LSTM models that consider only either forward or backward states, Bi-LSTM could account for temporal dependency of time-series images in both forward and backward directions. This property of Bi-LSTM can be effectively applied to crop classification when it is difficult to obtain full time-series images covering the entire growth cycle of crops. The classification performance of the Bi-LSTM is compared with that of two unidirectional LSTM architectures (forward and backward) with respect to different input image combinations via a case study of crop classification in Anbadegi, Korea. When full time-series images were used as inputs for classification, the Bi-LSTM outperformed the other unidirectional LSTM architectures; however, the difference in classification accuracy from unidirectional LSTM was not substantial. On the contrary, when using multitemporal images that did not include useful information for the discrimination of crops, the Bi-LSTM could compensate for the information deficiency by including temporal information from both forward and backward states, thereby achieving the best classification accuracy, compared with the unidirectional LSTM. These case study results indicate the efficiency of the Bi-LSTM for crop classification, particularly when limited input images are available.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.101-101
• /
• 2020

Traditional stochastic simulation of hydroclimatological variables often underestimates the variability and correlation structure of larger timescale due to the difficulty in preserving long-term memory. However, the Long Short-Term Memory (LSTM) model illustrates a remarkable long-term memory from the recursive hidden and cell states. The current study, therefore, employed the LSTM model in stochastic generation of hydrologic and climate variables to examine how much the LSTM model can preserve the long-term memory and overcome the drawbacks of conventional time series models such as autoregressive (AR). A trigonometric function and the Rössler system as well as real case studies for hydrological and climatological variables were tested. Results presented that the LSTM model reproduced the variability and correlation structure of the larger timescale as well as the key statistics of the original time domain better than the AR and other traditional models. The hidden and cell states of the LSTM containing the long-memory and oscillation structure following the observations allows better performance compared to the other tested conventional models. This good representation of the long-term variability can be important in water manager since future water resources planning and management is highly related with this long-term variability.

• Communications for Statistical Applications and Methods
• /
• v.26 no.5
• /
• pp.497-506
• /
• 2019

Forecasting the U.S. employment level is made using machine learning methods of the artificial neural network: deep neural network, long short term memory (LSTM), gated recurrent unit (GRU). We consider the big data of the federal reserve economic data among which 105 important macroeconomic variables chosen by McCracken and Ng (Journal of Business and Economic Statistics, 34, 574-589, 2016) are considered as predictors. We investigate the influence of the two statistical issues of the dimension reduction and time series differencing on the machine learning forecast. An out-of-sample forecast comparison shows that (LSTM, GRU) with differencing performs better than the autoregressive model and the dimension reduction improves long-term forecasts and some short-term forecasts.

• Advanced Industrial SCIence
• /
• v.3 no.1
• /
• pp.17-22
• /
• 2024

History, which began with the emergence of mankind, has a means of recording. Today, we can check the past through data. Generated data may only be generated and stored at a certain moment, but it is not only continuously generated over a certain time interval from the past to the present, but also occurs in the future, so making predictions using it is an important task. In order to find out trends in the use of time series data among numerous data, this paper analyzes the concept of time series data, analyzes Recurrent Neural Network and Long-Short Term Memory, which are mainly used for time series data analysis in the machine learning field, and analyzes the use of these models. Through case studies, it was confirmed that it is being used in various fields such as medical diagnosis, stock price analysis, and climate prediction, and is showing high predictive results. Based on this, we will explore ways to utilize it in the future.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.29 no.5
• /
• pp.505-511
• /
• 2023

In the present study, a deep learning model was established to predict the motion response of small fishing vessels. Hydrodynamic performances were evaluated for two small fishing vessels for the dataset of deep learning model. The deep learning model of the Long Short-Term Memory (LSTM) which is one of the recurrent neural network was utilized. The input data of LSTM model consisted of time series of six(6) degrees of freedom motions and wave height and the output label was selected as the time series data of six(6) degrees of freedom motions. The hyperparameter and input window length studies were performed to optimize LSTM model. The time series motion response according to different wave direction was predicted by establised LSTM. The predicted time series motion response showed good overall agreement with the analysis results. As the length of the time series increased, differences between the predicted values and analysis results were increased, which is due to the reduced influence of long-term data in the training process. The overall error of the predicted data indicated that more than 85% of the data showed an error within 10%. The established LSTM model is expected to be utilized in monitoring and alarm systems for small fishing vessels.

• Journal of Ocean Engineering and Technology
• /
• v.21 no.4
• /
• pp.28-33
• /
• 2007

The objective of this paper is to investigate the relationship between the short-term creep rupture time and the creep rupture properties at three different elevated temperatures in STS304 stainless steel. Uniaxial constant stress creep rupture tests were performed on the steel to observe the creep rupture behaviors at the elevated temperatures of 600, 650 and 700, according to the testing matrix. It is very important to predict creep life in practical creep design problems. As one of the series of studies on the statistical modelling of probabilistic creep rupture time and the development of creep life prediction techniques, the relationship between applied stress and creep rupture behaviors, such as creep strain rate and rupture time, were investigated. In addition, the Monkman-Grant relationship was observed between the steady-state creep rate and the creep rupture time. The creep rupture surfaces observed by SEM showed up dimple phenomenon at all conditions.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.2
• /
• pp.149-154
• /
• 2013

Short term load forecasting for electric power demand is essential for stable power system operation and efficient power market operation. High accuracy of the short term load forecasting can keep the power system more stable and save the power market operation cost. We propose an optimal coefficient selection method for exponential smoothing model in short term load forecasting on weekdays. In order to find the optimal coefficient of exponential smoothing model, load forecasting errors are minimized for actual electric load demand data of last three years. The proposed method are verified by case studies for last three years from 2009 to 2011. The results of case studies show that the average percentage errors of the proposed load forecasting method are improved comparing with errors of the previous methods.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.26 no.4
• /
• pp.363-368
• /
• 1993

Short-term forecasting of fish catch is of practical importance in fisheries management. Ecosystem models and multi-species models as well as traditional single-species models fall short of predicting power needed for practical management of fisheries resources due to the lack of sufficient data or information for the required parameters. Univariate time series analysis, on the other hand, extracts the information on the stochastic variability from the time series itself and makes estimates of the future stochastic variability. Therefore, it can be used for short-term forecasting with minimum data requirements. ARIMA time series modeling has been applied to the monthly Korean catches of hairtail (Trichiurus lepturus) for $1971{\sim}1988$. Forecasts of hairtail catch were made and compared with the actual catch data from $1989{\sim}1990$ which were not included in the parameter estimation. The results showed a good agreement (r=0.938) between the forecasts and the actual catches with a mean rotative error of $59.5\%$

• Journal of Ocean Engineering and Technology
• /
• v.30 no.3
• /
• pp.161-168
• /
• 2016

This paper presents a methodology through which the time series of the dynamic response of mooring line tension can be predicted without relying on a time-consuming nonlinear time-domain analysis. The mooring line tension for the target short-term sea states was predicted using a Hammerstein-Wiener model, a popular system identification scheme, based upon the pre-calculated motion-tension time history data for some selected short-term sea states that do not overlap with the targeted ones. The obtained mooring line tension was further processed, and a fatigue damage comparison was made between the predicted and calculated values. The results showed that the predicted time series of the mooring line tension matched the calculated one fairly well. Thus, it is expected that the methodology may be employed to enhance the efficiency of mooring line tension analysis.