• Journal of the Korean Society of Systems Engineering
• /
• v.16 no.1
• /
• pp.51-57
• /
• 2020

The fighter pilot uses radar mounted on the fighter to obtain high-resolution SAR (Synthetic Aperture Radar) images for a specific area of distance, and then the pilot visually classifies targets within the image. However, the target configuration captured in the SAR image is relatively small in size, and distortion of that type occurs depending on the depression angle, making it difficult for pilot to classify the type of target. Also, being present with various types of clutters, there should be errors in target classification and pilots should be even worse if tasks such as navigation and situational awareness are carried out simultaneously. In this paper, the concept of operation and functional structure of radar system for fighter jets were presented to transfer the SAR image target classification task of fighter pilots to radar system, and the method of target classification with high accuracy was studied using the CNN ensemble model to archive higher classification accuracy than single CNN model.

• Interaction and multiscale mechanics
• /
• v.4 no.3
• /
• pp.173-185
• /
• 2011

A multiscale modeling scheme that addresses the influence of the nanoparticle size in nanocomposites consisting of nano-sized spherical particles embedded in a polymer matrix is presented. A micromechanics-based constitutive model for nanoparticle-reinforced polymer composites is derived by incorporating the Eshelby tensor considering the interface effects (Duan et al. 2005a) into the ensemble-volume average method (Ju and Chen 1994). A numerical investigation is carried out to validate the proposed micromechanics-based constitutive model, and a parametric study on the interface moduli is conducted to investigate the effect of interface moduli on the overall behavior of the composites. In addition, molecular dynamics (MD) simulations are performed to determine the mechanical properties of the nanoparticles and polymer. Finally, the overall elastic moduli of the nanoparticle-reinforced polymer composites are estimated using the proposed multiscale approach combining the ensemble-volume average method and the MD simulation. The predictive capability of the proposed multiscale approach has been demonstrated through the multiscale numerical simulations.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.43 no.1
• /
• pp.123-131
• /
• 2020

There have been a lot of studies in the past for the method of predicting the failure of a machine, and recently, a lot of researches and applications have been generated to diagnose the physical condition of the machine and the parts and to calculate the remaining life through various methods. Survival models are also used to predict plant failures based on past anomaly cycles. In particular, special machine that reflect the fluid flow and process characteristics of chemical plants are connected to hundreds or thousands of sensors, so there are not many factors that need to be considered, such as process and material data as well as application of derivative variables. In this paper, the data were preprocessed through time series anomaly detection based on unsupervised learning to predict the abnormalities of these special machine. Next, clustering results reflecting clustering-based data characteristics were applied to produce additional variables, and a learning data set was created based on the history of past facility abnormalities. Finally, the prediction methodology based on the supervised learning algorithm was applied, and the model update was confirmed to improve the accuracy of the prediction of facility failure. Through this, it is expected to improve the efficiency of facility operation by flexibly replacing the maintenance time and parts supply and demand by predicting abnormalities of machine and extracting key factors.

• ETRI Journal
• /
• v.41 no.3
• /
• pp.371-382
• /
• 2019

Machine reading comprehension is the task of understanding a given context and finding the correct response in that context. A simple recurrent unit (SRU) is a model that solves the vanishing gradient problem in a recurrent neural network (RNN) using a neural gate, such as a gated recurrent unit (GRU) and long short-term memory (LSTM); moreover, it removes the previous hidden state from the input gate to improve the speed compared to GRU and LSTM. A self-matching network, used in R-Net, can have a similar effect to coreference resolution because the self-matching network can obtain context information of a similar meaning by calculating the attention weight for its own RNN sequence. In this paper, we construct a dataset for Korean machine reading comprehension and propose an $S^2-Net$ model that adds a self-matching layer to an encoder RNN using multilayer SRU. The experimental results show that the proposed $S^2-Net$ model has performance of single 68.82% EM and 81.25% F1, and ensemble 70.81% EM, 82.48% F1 in the Korean machine reading comprehension test dataset, and has single 71.30% EM and 80.37% F1 and ensemble 73.29% EM and 81.54% F1 performance in the SQuAD dev dataset.

• ETRI Journal
• /
• v.42 no.6
• /
• pp.886-898
• /
• 2020

Unusual data patterns or outliers can be generated because of human errors, incorrect measurements, or malicious activities. Detecting outliers is a difficult task that requires complex ensembles. An ideal outlier detection ensemble should consider the strengths of individual base detectors while carefully combining their outputs to create a strong overall ensemble and achieve unbiased accuracy with minimal variance. Selecting and combining the outputs of dissimilar base learners is a challenging task. This paper proposes a model that utilizes heterogeneous base learners. It adaptively boosts the outcomes of preceding learners in the first phase by assigning weights and identifying high-performing learners based on their local domains, and then carefully fuses their outcomes in the second phase to improve overall accuracy. Experimental results from 10 benchmark datasets are used to train and test the proposed model. To investigate its accuracy in terms of separating outliers from inliers, the proposed model is tested and evaluated using accuracy metrics. The analyzed data are presented as crosstabs and percentages, followed by a descriptive method for synthesis and interpretation.

• International Journal of Computer Science & Network Security
• /
• v.21 no.12spc
• /
• pp.526-538
• /
• 2021

Machine and deep learning-based models are emerging techniques that are being used to address prediction problems in biomedical data analysis. DNA sequence prediction is a critical problem that has attracted a great deal of attention in the biomedical domain. Machine and deep learning-based models have been shown to provide more accurate results when compared to conventional regression-based models. The prediction of the gene sequence that leads to cancerous diseases, such as prostate cancer, is crucial. Identifying the most important features in a gene sequence is a challenging task. Extracting the components of the gene sequence that can provide an insight into the types of mutation in the gene is of great importance as it will lead to effective drug design and the promotion of the new concept of personalised medicine. In this work, we extracted the exons in the prostate gene sequences that were used in the experiment. We built a Deep Neural Network (DNN) and Bi-directional Long-Short Term Memory (Bi-LSTM) model using a k-mer encoding for the DNA sequence and one-hot encoding for the class label. The models were evaluated using different classification metrics. Our experimental results show that DNN model prediction offers a training accuracy of 99 percent and validation accuracy of 96 percent. The bi-LSTM model also has a training accuracy of 95 percent and validation accuracy of 91 percent.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2021.06a
• /
• pp.205-205
• /
• 2021

Surface soil moisture, which governs the partitioning of precipitation into infiltration and runoff, plays an important role in the hydrological cycle. The assimilation of satellite soil moisture retrievals into a land surface model or hydrological model has been shown to improve the predictive skill of hydrological variables. This study aims to improve streamflow prediction with Weather Research and Forecasting model-Hydrological modeling system (WRF-Hydro) by assimilating Soil Moisture Active and Passive (SMAP) data at 3 km and analyze its impacts on hydrological components. We applied Cumulative Distribution Function (CDF) technique to remove the bias of SMAP data and assimilate SMAP data (April to July 2015-2019) into WRF-Hydro by using an Ensemble Kalman Filter (EnKF) with a total 12 ensembles. Daily inflow and soil moisture estimates of major dams (Soyanggang, Chungju, Sumjin dam) of South Korea were evaluated. We investigated how hydrologic variables such as runoff, evaporation and soil moisture were better simulated with the data assimilation than without the data assimilation. The result shows that the correlation coefficient of topsoil moisture can be improved, however a change of dam inflow was not outstanding. It may attribute to the fact that soil moisture memory and the respective memory of runoff play on different time scales. These findings demonstrate that the assimilation of satellite soil moisture retrievals can improve the predictive skill of hydrological variables for a better understanding of the water cycle.

• KIPS Transactions on Software and Data Engineering
• /
• v.9 no.6
• /
• pp.187-194
• /
• 2020

This paper proposes a website fingerprinting method using ensemble learning over a Tor network that guarantees client anonymity and personal information. We construct a training problem for website fingerprinting from the traffic packets collected in the Tor network, and compare the performance of the website fingerprinting system using tree-based ensemble models. A training feature vector is prepared from the general information, burst, cell sequence length, and cell order that are extracted from the traffic sequence, and the features of each website are represented with a fixed length. For experimental evaluation, we define four learning problems (Wang14, BW, CW_T, CW_H) according to the use of website fingerprinting, and compare the performance with the support vector machine model using CUMUL feature vectors. In the experimental evaluation, the proposed statistical-based training feature representation is superior to the CUMUL feature representation except for the BW case.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.2A
• /
• pp.95-104
• /
• 2012

Pattern recognition using ensemble classifiers is composed of strong classifier which consists of many weak classifiers. In this paper, we used feature extraction to organize strong classifier using static camera sequence. The strong classifier is made of weak classifiers which considers environmental factors. So the strong classifier overcomes environmental effect. Proposed method uses binary foreground image by frame difference method and the boosting is used to train crowdedness model and recognize crowdedness using features. Combination of weak classifiers makes strong ensemble classifier. The classifier could make use of potential features from the environment such as shadow and reflection. We tested the proposed system with road sequence and subway platform sequence which are included in "AVSS 2007" sequence. The result shows good accuracy and efficiency on complex environment.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.261-269
• /
• 2013

In this paper, the evoked potential(EP) was represented by additive delay model to comply with the variational noisy response of stimulus-event synchronization. The hybrid method of delay compensated-Wiener filtered-ensemble averaging(DWEA) was proposed to enhance the EP signal distortion occurred during averaging procedure due to synchronization timing mismatch. The performance of DWEA has been tested by surrogated simulation, which is composed of synthesized arbitrary delay and arbitrary level of added noise. The performance of DWEA is better than those of Wiener filtered-ensemble averaging and of conventional ensemble averaging. DWEA is endurable up to added noise gain of 7 for 10 % mean square error limit. Throughout the experimentation observation, it has been demonstrated that DWEA can be applied to enhance the evoked potential having the synchronization mismatch with added noise.