• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.1975-1988
• /
• 2019

Short-term traffic flow prediction plays an important role in intelligent transportation systems (ITS) in areas such as transportation management, traffic control and guidance. For short-term traffic flow regression predictions, the main challenge stems from the non-stationary property of traffic flow data. In this paper, we design an ensemble cascading prediction framework based on extremely randomized trees (extra-trees) using a boosting technique called EET to predict the short-term traffic flow under non-stationary environments. Extra-trees is a tree-based ensemble method. It essentially consists of strongly randomizing both the attribute and cut-point choices while splitting a tree node. This mechanism reduces the variance of the model and is, therefore, more suitable for traffic flow regression prediction in non-stationary environments. Moreover, the extra-trees algorithm uses boosting ensemble technique averaging to improve the predictive accuracy and control overfitting. To the best of our knowledge, this is the first time that extra-trees have been used as fundamental building blocks in boosting committee machines. The proposed approach involves predicting 5 min in advance using real-time traffic flow data in the context of inherently considering temporal and spatial correlations. Experiments demonstrate that the proposed method achieves higher accuracy and lower variance and computational complexity when compared to the existing methods.

• The Korean Journal of Applied Statistics
• /
• v.27 no.3
• /
• pp.357-371
• /
• 2014

There are many studies related to imbalanced data in which the class distribution is highly skewed. To address the problem of imbalanced data, previous studies deal with resampling techniques which correct the skewness of the class distribution in each sampled subset by using under-sampling, over-sampling or hybrid-sampling such as SMOTE. Ensemble methods have also alleviated the problem of class imbalanced data. In this paper, we compare around a dozen algorithms that combine the ensemble methods and resampling techniques based on simulated data sets generated by the Backbone model, which can handle the imbalance rate. The results on various real imbalanced data sets are also presented to compare the effectiveness of algorithms. As a result, we highly recommend the resampling technique combining ensemble methods for imbalanced data in which the proportion of the minority class is less than 10%. We also find that each ensemble method has a well-matched sampling technique. The algorithms which combine bagging or random forest ensembles with random undersampling tend to perform well; however, the boosting ensemble appears to perform better with over-sampling. All ensemble methods combined with SMOTE outperform in most situations.

• Journal of Korea Water Resources Association
• /
• v.51 no.3
• /
• pp.221-233
• /
• 2018

Recently, the flood damage by the localized heavy rainfall and typhoon have been frequently occurred due to the climate change. Accurate rainfall forecasting and flood runoff estimates are needed to reduce such damages. However, the uncertainties are involved in guage rainfall, radar rainfall, and the estimated runoff hydrograph from rainfall-runoff models. Therefore, the purpose of this study is to identify the uncertainty of rainfall by generating a probabilistic radar rainfall ensemble and confirm the uncertainties of hydrological models through the analysis of the simulated runoffs from the models. The blending technique is used to estimate a single integrated or an optimal runoff hydrograph by the simulated runoffs from multi rainfall-runoff models. The radar ensemble is underestimated due to the influence of rainfall intensity and topography and the uncertainty of the rainfall ensemble is large. From the study, it will be helpful to estimate and predict the accurate runoff to prepare for the disaster caused by heavy rainfall.

• Proceedings of the KOSOMBE Conference
• /
• v.1989 no.05
• /
• pp.71-73
• /
• 1989

The origins of the motion artifact resulting from exercise in impedance cardiography wore explained and the ensemble average technique was applied to reduce the motion artifact enabling the measurement of cardiac output during exercise. Algorithm for ensemble average was developed and applied to the actual impedance signals. It was found that the minimum number of sampling was 20, and sampling frequency was 500Hz. Using the ensemble average technique it was possible to measure cardiac output continuously during the treadmill exercise. Therefore it is hoped that this study may contribute in the area of exercise physiology and sport medicine.

• Proceedings of the KOSOMBE Conference
• /
• v.1994 no.12
• /
• pp.124-127
• /
• 1994

A technique based on bispectrun averaging is described for generally recovering the signal waveform from a set of noisy signals with variable signal delay. The technique does not require explicit tune alignment of signals and any initial estimate of signal. The new method is suggested and is compared with other methods. This method are numerically investigated using computer generated-data and a physiological signal and noise Some experimental results for the evoked potential studios that demonstrate the technique are given. The results show the effectiveness of the technique: various potential applications of the technique might be expected.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.58 no.8
• /
• pp.1638-1643
• /
• 2009

Uroflowmetry is a convenient clinical test to screen the benign prostatic hyperplasia(BPH) common in the aged men. A load cell is located beneath the urine container to measure the weight of urine. However, it is sensitive to the impact applied on the bottom of the container by the urine stream, which could be a noise source lowering the reliability of the system. With this aim, our study proposed a noise reduction technique by computing ensemble average of the weighted signals that were acquired from three-load cells forming a regular triangle beneath the urine container. Simulated urination experiment was performed with three different collection methods, all of which demonstrated significant noise reduction by ensemble averaging. Furthermore, the best results can be obtained without any special urine collection devices. Thus, our novel method can be usefully applied to uroflowmetry for enhancing measurement in terms of accuracy and reliability.

• Proceedings of the KOSOMBE Conference
• /
• v.1990 no.05
• /
• pp.7-8
• /
• 1990

In this study, a new ensemble average technique was developed to measure cardiac output during treadmill exercise. Each dZ/dt peak (C point) was used as a starting point for ensemble averaging, instead of conventionally used R wave of ECG in order to prevent the peak dZ/dt waveform from blurring. In ease of using R wave as a reference, time interval from R wave to the peak of dZ/dt varies for each heart beat. Stroke volume, heart rate, and cardiac output of five male were successfully measured with Balke protocol using the new ensemble average technique.

• Journal of Digital Convergence
• /
• v.17 no.4
• /
• pp.115-122
• /
• 2019

In the online marketplace, consumers are exposed to various products and freely express opinions. As consumer product reviews have a important effect on the success of online markets and other consumers, online market needs to accurately analyze the consumers' emotions about their products. Text mining, which is one of the data analysis techniques, can analyze the consumer's reviews on the products and efficiently manage the products. Previous studies have analyzed specific domains and less than 20,000 data, despite the different accuracy of the analysis results depending on the data domain and size. Further, there are few studies on additional factors that can improve the accuracy of analysis. This study analyzed 72,530 review data of food product domain that was not mainly covered in previous studies by using ensemble technique. We also examined the influence of summary review on improving accuracy of analysis. As a result of the study, this study found that Boosting ensemble technique has the highest accuracy of analysis. In addition, the summary review contributed to improving accuracy of the analysis.

• Journal of Korea Society of Industrial Information Systems
• /
• v.17 no.7
• /
• pp.139-148
• /
• 2012

An ensemble of classifiers is to employ a set of individually trained classifiers and combine their predictions. It has been found that in most cases the ensembles produce more accurate predictions than the base classifiers. Combining outputs from multiple classifiers, known as ensemble learning, is one of the standard and most important techniques for improving classification accuracy in machine learning. An ensemble of classifiers is efficient only if the individual classifiers make decisions as diverse as possible. Bagging is the most popular method of ensemble learning to generate a diverse set of classifiers. Diversity in bagging is obtained by using different training sets. The different training data subsets are randomly drawn with replacement from the entire training dataset. The random subspace method is an ensemble construction technique using different attribute subsets. In the random subspace, the training dataset is also modified as in bagging. However, this modification is performed in the feature space. Bagging and random subspace are quite well known and popular ensemble algorithms. However, few studies have dealt with the integration of bagging and random subspace using SVM Classifiers, though there is a great potential for useful applications in this area. The focus of this paper is to propose methods for improving SVM performance using hybrid ensemble strategy for bankruptcy prediction. This paper applies the proposed ensemble model to the bankruptcy prediction problem using a real data set from Korean companies.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.25-25
• /
• 2023

The ensemble optimal interpolation (EnOI) scheme is a sub-optimal alternative to the ensemble Kalman filter (EnKF) with a reduced computational demand making it potentially more suitable for operational applications. Since only one model is integrated forward instead of an ensemble of model realizations, online estimation of the background error covariance matrix is not possible in the EnOI scheme. In this study, we investigate two Gaussian noise based ensemble generation strategies to produce dynamic covariance matrices for assimilation of water level observations into a distributed hydrological model. In the first approach, spatially correlated noise, sampled from a normal distribution with a fixed fractional error parameter (which controls its standard deviation), is added to the model forecast state vector to prepare the ensembles. In the second method, we use an adaptive error estimation technique based on the innovation diagnostics to estimate this error parameter within the assimilation framework. The results from a real and a set of synthetic experiments indicate that the EnOI scheme can provide better results when an optimal EnKF is not identified, but performs worse than the ensemble filter when the true error characteristics are known. Furthermore, while the adaptive approach is able to reduce the sensitivity to the fractional error parameter affecting the first (non-adaptive) approach, results are usually worse at ungauged locations with the former.