• Transactions on Control, Automation and Systems Engineering
• /
• v.1 no.2
• /
• pp.113-120
• /
• 1999

One of the important challenges facing control engineers in developing automated machineryis to be able to monitor the machines using remote sensors. Observrs are often used to reconstruct the machine variables of interest. However, conventional observers are unalbe to observe the machine variables when the machine models, upon which the observers are based, have inputs that cannot be measured. Since this is often the case, the authors previsously developed a steady-state optimal state and input observer for time-invariant systems [1], this paper extends that work to time-variant systems. A recursive observer, similar to a Kalman-Bucy filter, is developed . This optimal observer minimizes the trace of the error variance for discrete , linear , time-variant, stochastic systems with unknown inputs.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.279-280
• /
• 2008

In order to improve the performance and simplify the structure of the conventional detectors in high density magnetic channels, a new equalizer based on bilinear recursive polynomial (BRP) models, which uses the previously estimated sequence, is proposed. The performance is compared with the conventional equalizers and the maximum likelihood sequence detection (MLSD) bound.

• Journal of the Korean Statistical Society
• /
• v.26 no.2
• /
• pp.277-288
• /
• 1997

For the analysis of survival data including covariates whose effects vary in time, the multiprocess discount survival model is proposed. The parameter vector modeling the time-varying effects of covariates is to vary between time intervals and its evolution between time intervals depends on the perturbation of the next time interval. The recursive estimation of the parameter vector can be obtained at the end of each time interval. The retrospective estimation of the survival function and the forecasting of the survival function of individuals of the specific covariates also can be obtained based on the information gathered until the end of the time interval.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.24 no.4
• /
• pp.286-293
• /
• 2019

This study compares and analyzes online update techniques, which estimate the parameters of battery equivalent circuit models in real time. Online update techniques, which are based on extended Kalman filter and recursive least square methods, are constructed by considering the dynamic characteristics of batteries. The performance of the online update techniques is verified by simulation and experiments. Each online update technique is compared and analyzed in terms of complexity and accuracy to propose a suitable guide for selecting algorithms on various types of battery applications.

• Journal of information and communication convergence engineering
• /
• v.21 no.1
• /
• pp.82-89
• /
• 2023

Various machine-learning models may yield high predictive power for massive time series for time series prediction. However, these models are prone to instability in terms of computational cost because of the high dimensionality of the feature space and nonoptimized hyperparameter settings. Considering the potential risk that model training with a high-dimensional feature set can be time-consuming, we evaluate a feature-importance-based feature selection method to derive a tradeoff between predictive power and computational cost for time series prediction. We used two machine learning techniques for performance evaluation to generate prediction models from a retail sales dataset. First, we ranked the features using impurity- and Local Interpretable Model-agnostic Explanations (LIME) -based feature importance measures in the prediction models. Then, the recursive feature elimination method was applied to eliminate unimportant features sequentially. Consequently, we obtained a subset of features that could lead to reduced model training time while preserving acceptable model performance.

• The Journal of Engineering Geology
• /
• v.23 no.2
• /
• pp.137-147
• /
• 2013

For the effective management of groundwater resources, it is necessary to predict groundwater level fluctuations in response to rainfall events. In the present study, time series models using artificial neural networks (ANNs) and support vector machines (SVMs) have been developed and applied to groundwater level data from the Gasan, Shingwang, and Cheongseong stations of the National Groundwater Monitoring Network. We designed four types of model according to input structure and compared their performances. The results show that the rainfall input model is not effective, especially for the prediction of groundwater recession behavior; however, the rainfall-groundwater input model is effective for the entire prediction stage, yielding a high model accuracy. Recursive prediction models were also effective, yielding correlation coefficients of 0.75-0.95 with observed values. The prediction errors were highest for Shingwang station, where the cross-correlation coefficient is lowest among the stations. Overall, the model performance of SVM models was slightly higher than that of ANN models for all cases. Assessment of the model parameter uncertainty of the recursive prediction models, using the ratio of errors in the validation stage to that in the calibration stage, showed that the range of the ratio is much narrower for the SVM models than for the ANN models, which implies that the SVM models are more stable and effective for the present case studies.

• The Journal of Information Systems
• /
• v.26 no.4
• /
• pp.17-37
• /
• 2017

Purpose Customer Loyalty is the most important factor of customer relationship management (CRM). Especially in retailing industry, where customers have many options of where to spend their money. Classifying loyal customers through customers' data can help retailing companies build more efficient marketing strategies and gain competitive advantages. This study aims to construct classification models of distinguishing the loyal customers within a Korean retailing company using data mining techniques with R language. Design/methodology/approach In order to classify retailing customers, we used combination of support vector machines (SVMs) and other classification algorithms of machine learning (ML) with the support of recursive feature elimination (RFE). In particular, we first clean the dataset to remove outlier and impute the missing value. Then we used a RFE framework for electing most significant predictors. Finally, we construct models with classification algorithms, tune the best parameters and compare the performances among them. Findings The results reveal that ML classification techniques can work well with CRM data in Korean retailing industry. Moreover, customer loyalty is impacted by not only unique factor such as net promoter score but also other purchase habits such as expensive goods preferring or multi-branch visiting and so on. We also prove that with retailing customer's dataset the model constructed by SVMs algorithm has given better performance than others. We expect that the models in this study can be used by other retailing companies to classify their customers, then they can focus on giving services to these potential vip group. We also hope that the results of this ML algorithm using R language could be useful to other researchers for selecting appropriate ML algorithms.

• International Area Studies Review
• /
• v.21 no.4
• /
• pp.159-172
• /
• 2017

In a typical trend-cycle decomposition of GDP, the trend component is usually assumed to follow a random walk process. This paper considers an ARIMA trend and assesses the validity of the ARIMA trend model. I construct univariate and bivariate unobserved-components(UC) models, allowing the ARIMA trend. Estimation results using U.S. data are favorable to the ARIMA trend models. I, also, compare the forecasting performance of the UC models. Dynamic pseudo-out-of-sample forecasting exercises are implemented with recursive estimations. I find that the bivariate model outperforms the univariate model, the smoothed estimates of trend and cycle components deliver smaller forecasting errors compared to the filtered estimates, and, most importantly, allowing for the ARIMA trend can lead to statistically significant gains in forecast accuracy, providing support for the ARIMA trend model. It is worthy of notice that trend shocks play the main source of the output fluctuation if the ARIMA trend is allowed in the UC model.

• Bulletin of the Korean Chemical Society
• /
• v.33 no.2
• /
• pp.613-619
• /
• 2012

We applied several machine learning methods for developing QSAR models for prediction of acute toxicity to fathead minnow. The multiple linear regression (MLR) and artificial neural network (ANN) method were applied to predict 96 h $LC_{50}$ (median lethal concentration) of 555 chemical compounds. Molecular descriptors based on 2D chemical structure were calculated by PreADMET program. The recursive partitioning (RP) model was used for grouping of mode of actions as reactive or narcosis, followed by MLR method of chemicals within the same mode of action. The MLR, ANN, and two RP-MLR models possessed correlation coefficients ($R^2$) as 0.553, 0.618, 0.632, and 0.605 on test set, respectively. The consensus model of ANN and two RP-MLR models was used as the best model on training set and showed good predictivity ($R^2$=0.663) on the test set.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.6
• /
• pp.728-734
• /
• 1999

An on-line system identification scheme of unknown system is proposed based on a Laguerre models representation. The unknown parameters are detemined using recursive least-square identification. The proposed method have the advantage that an unknown system can be modelled without structural knowledge and assumption about the true model order and time delay. Therefore, the proposed method can make the design procedure very when compared to widely-used conventional method.