• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 1999.11a
• /
• pp.216-221
• /
• 1999

A mental task combined with noise background is an effective model of laboratory stress for study of psychophysiology of the autonomic nervous system (ANS). The intensity of the background noise significantly affects both a subjective evaluation of experienced stress level during test and the physiological responses associated with mental load in noisy environments. Providing tests of similar difficulties we manipulated the background noise intensity as a main factor influencing a psychophysiological outcome and the analyzed reactivity along withe the noise intensity dimension. The goal of this study was to identify the patterns of ANS responses and the relevant subjective stress scores during performance of word recognition tasks on the background of white noise (WN) of the different intensities (55, 70 and 85 dB). Subjects were 27 college students (19-24 years old). BIOPAC, Grass Neurodata System and AcqKnowlwdge 3.5 software were used to record ECG, PPG, SCL, skin temperature, and respiration. Experimental manipulations were effective in producing subjective and physiological responses usually associated with stress. The results suggested that the following potential autonomic mechanisms might be involved in the mediation of the observed physiological responses: A sympathetic activation with parasympathetic withdrawal during mild 55 and 70dB noise (featured by similar profiles) and simultaneous activation of sympathetic and parasympathetic systems during intense 85dB WN. The parasympathetic activation in this case might be a compensatory effect directed to prevent sympathetic domination and to maintain optimal arousal state for the successful performance on mental stress task. It should be mentioned that obtained results partially support Gellhorn's (1960; 1970) "tuning phenomenon" as a possible mechanism underlying stress response.

• The KIPS Transactions:PartD
• /
• v.16D no.2
• /
• pp.191-200
• /
• 2009

Location based services (LBSs) have drawn huge attention as the growing number mobile devices and wireless technology demand more pervasive information access. In LBSs, Constraint nearest neighbor queries are one of the important queries of human to fulfill his desired quickly through wireless technology. We propose new query processing of constrained nearest neighbor query on the air to support mobile clients which demand optimal uses of wireless broadcast channel and using minimum battery power of client. First we proposed NN query processing with constrained region and then explain the novel NN query processing with various types of constraints. We have proposed novel algorithms to support Constrained Nearest Neighbor queries on the air based on Distributed Spatial Index and Bitmap-based Spatial Index.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.3
• /
• pp.360-366
• /
• 2008

The initial fuzzy partitions in fuzzy rule-based classification systems are determined by considering the domain region of each attribute with the given data, and the optimal classification boundaries within the fuzzy partitions can be discovered by tuning their parameters using various learning processes such as neural network, genetic algorithm, and so on. In this paper, we propose a selection method for fuzzy partition based on statistical information to maximize the performance of pattern classification without learning processes where statistical information is used to extract the uncertainty regions (i.e., the regions which the classification boundaries in pattern classification problems are determined) in each input attribute from the numerical data. Moreover the methods for extracting the candidate rules which are associated with the partition intervals generated by statistical information and for minimizing the coupling problem between the candidate rules are additionally discussed. In order to show the effectiveness of the proposed method, we compared the classification accuracy of the proposed with those of conventional methods on the IRIS and New Thyroid Cancer data. From experimental results, we can confirm the fact that the proposed method only considering statistical information of the numerical patterns provides equal to or better classification accuracy than that of the conventional methods.

• Journal of the Korea Convergence Society
• /
• v.10 no.5
• /
• pp.165-171
• /
• 2019

Engineering and Technology have been influencing a lot in the field of sports. Competitiveness, attributes of sports, have forced not only sports players but sports goods to enhance those performance. Particularly in the field of motorsports, the convergence of sports and technology has long been done to satisfy between performance and safety. In this study, strength analysis was carried with FEM to develop CFRP Laminate(Carbon Fiber Reinforced Plastic Laminate) bucket seat targeted to motorsports and car tuning industries and FIA($F\acute{e}d\acute{e}ration$ Internationale de l'Automobile) regulation was applied to design the racing seat and evaluate its strength. FEM modeling considered the attributes of composites was followed by strength evaluation based on Tsai-Wu failure index were done according to Lay-up sequence and layer numbers. The result showed that the lay-up sequence with stacking angle such as $[0^{\circ}/30^{\circ}/60^{\circ}/90^{\circ}/-30^{\circ}/-60^{\circ}]_4$ with 3mm form core was optimal selection in the field of weight and strength evaluation.

• Korean Journal of Remote Sensing
• /
• v.35 no.2
• /
• pp.279-288
• /
• 2019

We evaluated the land cover classification performance of SegNet, which features semantic segmentation of aerial imagery. We selected four semantic classes, i.e., urban, farmland, forest, and water areas, and created 2,000 datasets using aerial images and land cover maps. The datasets were divided at a 8:2 ratio into training (1,600) and validation datasets (400); we evaluated validation accuracy after tuning the hyperparameters. SegNet performance was optimal at a batch size of five with 100,000 iterations. When 200 test datasets were subjected to semantic segmentation using the trained SegNet model, the accuracies were farmland 87.89%, forest 87.18%, water 83.66%, and urban regions 82.67%; the overall accuracy was 85.48%. Thus, deep learning-based semantic segmentation can be used to classify land cover.

• Journal of the Korea Convergence Society
• /
• v.12 no.1
• /
• pp.49-55
• /
• 2021

Recently, with the development of database, it is possible to store a lot of data generated in finance, security, and networks. These data are being analyzed through classifiers based on machine learning. The main problem at this time is data imbalance. When we train imbalanced data, it may happen that classification accuracy is degraded due to over-fitting with majority class data. To overcome the problem of data imbalance, oversampling strategy that increases the quantity of data of minority class data is widely used. It requires to tuning process about suitable method and parameters for data distribution. To improve the process, In this study, we propose a strategy to explore and optimize oversampling combinations and ratio based on various methods such as synthetic minority oversampling technique and generative adversarial networks through genetic algorithms. After sampling credit card fraud detection which is a representative case of data imbalance, with the proposed strategy and single oversampling strategies, we compare the performance of trained classifiers with each data. As a result, a strategy that is optimized by exploring for ratio of each method with genetic algorithms was superior to previous strategies.

• Korean Journal of Materials Research
• /
• v.31 no.3
• /
• pp.115-121
• /
• 2021

MnO2 can be potentially utilized as an electrode material for redox capacitors. The deposition of MnO2 with poor electrical conductivity onto porous carbons supplies them with additional conductive paths; as a result, the capacitance of the electrical double layer formed on the porous carbon surface can be utilized together with the redox capacitance of MnO2. However, the obtained composites are not generally suitable for industrial production because they require the use of expensive porous carbons and/or inefficient fabrication methods. Thus, to develop an effective preparation procedure of the composite, a suitable structure of porous carbons must be determined. In this study, MnO2/C composites have been prepared from activated carbon gels with various pore sizes, and their electrical properties are investigated via cyclic voltammetry. In particular, mesoporous carbons with a pore size of around 20 nm form a composite with a relatively low capacitance (98 F/g-composite) and poor rate performance despite the moderate redox capacitance obtained for MnO2 (313 F/g-MnO2). On the other hand, using macro-porous carbons with a pore size of around 60 nm increases the MnO2 redox capacitance (399 F/g-MnO2) as well as the capacitance and rate performance of the entire material (203 F/g-composite). The obtained results can be used in the industrial manufacturing of MnO2/C composites for supercapacitor electrodes from the commercially available porous carbons.

• Journal of Digital Convergence
• /
• v.20 no.4
• /
• pp.459-468
• /
• 2022

In medical area, very high resolution images, which is loss sensitive data, are used. Therefore, the use of optical internet with high bandwidth and the transmission of high realiability is required. However, according to the nature of the Internet, various data use the same bandwidth and a new scheme is needed to differentiate effectively these data. In order to achieve the differentiation, optical delay line buffers are used. However, these buffers is constructed based on some optimal values such as the average offered load, measured data burst length, and basic delay unit. Once the buffers are installed, they are impossible to reinstall new buffers. So, the scheme changing burst length dynamically was considered. However, this method is highly unstable. Therefore, in this article, in order to guarantee the stable operation of the scheme, the analysis of operation conditions is performed. With the analysis together with the scheme, high resolution medical data with the higher class can transmit stably without loss.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.36 no.4
• /
• pp.233-242
• /
• 2023

The unscented Kalman filter (UKF), which is widely used to estimate the states of nonlinear dynamic systems, can be improved to realize robust system identification by using multiple sigma-point sets. When using Kalman filter methods for system identification, artificial noises must be appropriately selected to achieve optimal estimation performance. Additionally, an appropriate scaling factor for the sigma-points must be selected to capture the nonlinearity of the state-space model. This study entailed the use of Bouc-Wen hysteresis model to examine the nonlinear behavior of a single-degree-of-freedom oscillator. On the basis of the effects of the selected artificial noises and scaling factor, a new UKF method using multiple sigma-point sets was devised for improved robustness of the estimation over various signal-to-noise-ratio values. The results demonstrate that the proposed method can accurately track nonlinear system states even when the measurement noise levels are high, while being robust to the selection of artificial noise levels.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.297-299
• /
• 2021

Predicting future productivity of tight oil is an important task for analyzing residual oil recovery and reservoir behavior. In general, productivity prediction is made using the decline curve analysis(DCA). In this study, we intend to propose an effective model for predicting future production using deep learning-based recurrent neural networks(RNN), LSTM, and GRU algorithms. As input variables, the main parameters are oil, gas, water, which are calculated during the production of tight oil, and the type curve calculated through various cluster analyzes. the output variable is the monthly oil production. Existing empirical models, the DCA and RNN models, were compared, and an optimal model was derived through hyperparameter tuning to improve the predictive performance of the model.