• KIPS Transactions on Computer and Communication Systems
• /
• v.12 no.2
• /
• pp.53-60
• /
• 2023

Recently, the demand and traffic volume for various multimedia contents are rapidly increasing through real-time streaming platforms. In this paper, we predict real-time streaming traffic to improve the quality of service (QoS). Statistical models have been used to predict network traffic. However, since real-time streaming traffic changes dynamically, we used recurrent neural network-based deep learning models rather than a statistical model. Therefore, after the collection and preprocessing for real-time streaming data, we exploit vanilla RNN, LSTM, GRU, Bi-LSTM, and Bi-GRU models to predict real-time streaming traffic. In evaluation, the training time and accuracy of each model are measured and compared.

• The Korean Journal of Applied Statistics
• /
• v.21 no.5
• /
• pp.825-833
• /
• 2008

Discriminant analysis based on Gaussian mixture models, an useful tool for multi-class classifications, can be extended to semi-supervised learning. We consider a model selection problem for a Gaussian mixture model in semi-supervised learning. More specifically, we adopt Bayesian information criterion to determine the number of subclasses in the mixture model. Through simulations, we illustrate the usefulness of the criterion.

• Journal of Korea Multimedia Society
• /
• v.22 no.10
• /
• pp.1133-1141
• /
• 2019

Most of the previous studies for bark recognition have focused on the extraction of LBP-like statistical features. Deep learning approach was not well studied because of the difficulty of acquiring large volume of bark image dataset. To overcome the bark dataset problem, this study utilizes the MobileNet which was trained with the ImageNet dataset. This study proposes two approaches. One is to extract features by the pixel-wise convolution and classify the features with SVM. The other is to tune the weights of the MobileNet by flexibly freezing layers. The experimental results with two public bark datasets, BarkTex and Trunk12, show that the proposed methods are effective in bark recognition. Especially the results of the flexible tunning method outperform state-of-the-art methods. In addition, it can be applied to mobile devices because the MobileNet is compact compared to other deep learning models.

• Journal of radiological science and technology
• /
• v.45 no.1
• /
• pp.41-47
• /
• 2022

The purpose of this study is to apply a deep learning model that can distinguish lung perfusion and lung ventilation images in nuclear medicine, and to evaluate the image classification ability. Image data pre-processing was performed in the following order: image matrix size adjustment, min-max normalization, image center position adjustment, train/validation/test data set classification, and data augmentation. The convolutional neural network(CNN) structures of VGG-16, ResNet-18, Inception-ResNet-v2, and SE-ResNeXt-101 were used. For classification model evaluation, performance evaluation index of classification model, class activation map(CAM), and statistical image evaluation method were applied. As for the performance evaluation index of the classification model, SE-ResNeXt-101 and Inception-ResNet-v2 showed the highest performance with the same results. As a result of CAM, cardiac and right lung regions were highly activated in lung perfusion, and upper lung and neck regions were highly activated in lung ventilation. Statistical image evaluation showed a meaningful difference between SE-ResNeXt-101 and Inception-ResNet-v2. As a result of the study, the applicability of the CNN model for lung scintigraphy classification was confirmed. In the future, it is expected that it will be used as basic data for research on new artificial intelligence models and will help stable image management in clinical practice.

• Journal of Oriental Neuropsychiatry
• /
• v.10 no.1
• /
• pp.17-38
• /
• 1999

The effects of Sahyangsohapwon on the affective reactivity of rats were studied with open-field behavior. Sample group was treated with the medicine for 8 weeks, whereas control group was treated with the vehicle. The effects of Sahyangsohapwon on the enhancement of learning and memory of AD model rats were studied with two-way avoidance task. Sample group electrically lesioned on nbM(nucleus basalis of Meynert) was treated with the medicine for 8 weeks, whereas control group with nbM lesion and sham group with the sham operation were treated with the vehicle. 1. In the open-field behavior task, the start latency from start box was measured $27.08{\pm}7.51sec$ in control group, $23.15{\pm}5.98sec$ in sample group. Rats in sample group showed a tendency of shortened latency going out to a strange place compared with those in control group, but with no statistical significance(p>0.05). 2. In the open-field behavior task, the number of locomotion crossing the grid lines was measured $84.54{\pm}3.55$ in control group, $116.93{\pm}6.41$ in sample group. There was an increased locomotion in sample group compared with control group with statistical significance(p<0.01). This can be interpreted as rats in sample group showed lowerd anxiety under a strange environment. 3. In the open-field behavior task, the rearing number was measured $7.46{\pm}0.57$ in control group, $10.13{\pm}0.95$ in sample group. There was an increased rearing in sample group compared with control group with statistical significance(p<0.05). This can also be interpreted as rats in sample group showed lowerd anxiety under a strange environment. 4. In the open-field behavior task, the number of crossing behavior was measured $5.54{\pm}1.50$ in control group, $9.20{\pm}1.67$ in sample group. There was a increasing tendency of crossing behavior in sample group compared with control group, but with no statistical significance(p<0.05). 5. In the open-field behavior task, the total activity was measured $97.54{\pm}4.70$ in control group, $136.27{\pm}792$ in sample group. There was an increased total activity in sample group compared with control group with statistical significance(p<0.01). This can also be interpreted as rats in sample group showed lowerd anxiety under a strange environment. 6. In the analysis of effects on the learning and memory in AD model rats with two-way avoidance task, the response latency was measured $6717{\pm}134msec$ in the 1st session, $5416{\pm}160msec$ in the 2nd session, $5252{\pm}148msec$ in the 3rd session in control group. It was measured $6724{\pm}155msec$ in the 1st session, $4642{\pm}139msec$ in the 2nd session, $4914{\pm}148msec$ in the 3rd session in sample group and $4357{\pm}144msec$ in the 1st session, $3125{\pm}115msec$ in the 2nd session, $3091{\pm}98msec$ in the 3rd session in sham group. There were differences between sham group and nbM lesioned groups with statistical significance in post hoc analysis(p<0.000). And in the 2nd session, there was a reduction of latency in sample group compared with control group with statistical significance (p<0.000). This showed that sample group had better learning capacity than control group. 7. In the analysis of effects on the learning and memory in AD model rats with two-way avoidance task, the number of avoidance response was measured $5.85{\pm}1.41$ in the 1st session, $14.23{\pm}2.89$ in the 2nd session, $15.69{\pm}2.56$ in the 3rd session in control group. It was measured $7.92{\pm}1.94$ in the 1st session, $16.83{\pm}2.29$ in the 2nd session, $15.42{\pm}2.81$ in the 3rd session in sample group and $14.38{\pm}1.62$ in the 1st session, $22.88{\pm}0.89$ in the 2nd session, $23.88{\pm}1.64$ in the 3rd session in sham group. There were differences between sham group and nbM lesioned groups with statistical significance in post hoc analysis(p<0.001). But between control and sample group, there was no significant difference. With the experimental results above, Sahyangsohapwon can be supposed to have the enhancing effects on the affect reactivity and learning with memory of AD model rats induced by electrolyte injury of nbM.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.6
• /
• pp.26-36
• /
• 2021

In this study, a deep learning model for predicting the queue length was developed using the information collected from the image detector. Then, a multiple regression analysis model, a statistical technique, was derived and compared using two indices of mean absolute error(MAE) and root mean square error(RMSE). From the results of multiple regression analysis, time, day of the week, occupancy, and bus traffic were found to be statistically significant variables. Occupancy showed the most strong impact on the queue length among the variables. For the optimal deep learning model, 4 hidden layers and 6 lookback were determined, and MAE and RMSE were 6.34 and 8.99. As a result of evaluating the two models, the MAE of the multiple regression model and the deep learning model were 13.65 and 6.44, respectively, and the RMSE were 19.10 and 9.11, respectively. The deep learning model reduced the MAE by 52.8% and the RMSE by 52.3% compared to the multiple regression model.

• Journal of Biomedical Engineering Research
• /
• v.43 no.6
• /
• pp.441-447
• /
• 2022

In this paper, we propose a pill identification model using engraved text feature and image feature such as shape and color, and compare it with an identification model that does not use engraved text feature to verify the possibility of improving identification performance by improving recognition rate of the engraved text. The data consisted of 100 classes and used 10 images per class. The engraved text feature was acquired through Keras OCR based on deep learning and 1D CNN, and the image feature was acquired through 2D CNN. According to the identification results, the accuracy of the text recognition model was 90%. The accuracy of the comparative model and the proposed model was 91.9% and 97.6%. The accuracy, precision, recall, and F1-score of the proposed model were better than those of the comparative model in terms of statistical significance. As a result, we confirmed that the expansion of the range of feature improved the performance of the identification model.

• Computers and Concrete
• /
• v.33 no.2
• /
• pp.137-145
• /
• 2024

The technique of experimentally determining concrete's compressive strength for a given mix design is time-consuming and difficult. The goal of the current work is to propose a best working predictive model based on different machine learning algorithms such as Gradient Boosting Machine (GBM), Stacked Ensemble (SE), Distributed Random Forest (DRF), Extremely Randomized Trees (XRT), Generalized Linear Model (GLM), and Deep Learning (DL) that can forecast the compressive strength of ternary geopolymer concrete mix without carrying out any experimental procedure. A geopolymer mix uses supplementary cementitious materials obtained as industrial by-products instead of cement. The input variables used for assessing the best machine learning algorithm not only include individual ingredient quantities, but molarity of the alkali activator and age of testing as well. Myriad statistical parameters used to measure the effectiveness of the models in forecasting the compressive strength of ternary geopolymer concrete mix, it has been found that GBM performs better than all other algorithms. A sensitivity analysis carried out towards the end of the study suggests that GBM model predicts results close to the experimental conditions with an accuracy between 95.6 % to 98.2 % for testing and training datasets.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2021.05a
• /
• pp.151-152
• /
• 2021

The sale price of apartment buildings is a key factor in the success or failure of apartment projects, and the factors that affect the sale price of apartments vary widely, including location, environmental factors, and economic conditions. Existing methods of predicting the sale price do not reflect the nonlinear characteristics of apartment prices, which are determined by the complex impact factors of reality, because statistical analysis is conducted under the assumption of a linear model. To improve these problems, a new analysis technique is needed to predict apartment sales prices by complex nonlinear influencing factors. Using machine learning techniques that have recently attracted attention in the field of engineering, it is possible to predict the sale price reflecting the complexity of various factors. Therefore, this study aims to conduct a basic study for the development of a machine learning-based prediction model for apartment sale prices.

• Journal of the Korean Data and Information Science Society
• /
• v.15 no.2
• /
• pp.477-483
• /
• 2004

One of the major paradigms for supervised learning in neural network community is back-propagation learning. The standard implementations of back-propagation learning are optimal under the assumptions of identical and independent Gaussian noise. In this paper, for regression function estimation, we introduce $\varepsilon-insensitive$ back-propagation learning algorithm, which corresponds to minimizing the least absolute error. We compare this algorithm with support vector machine(SVM), which is another $\varepsilon-insensitive$ supervised learning algorithm and has been very successful in pattern recognition and function estimation problems. For comparison, we consider a more realistic model would allow the noise variance itself to depend on the input variables.