• The Korean Journal of Community Living Science
• /
• v.28 no.1
• /
• pp.93-101
• /
• 2017

This study was conducted to assess the quality characteristics and antioxidant activity of yanggaeng prepared with different concentrations of Viscum album extracts (VAE; 0, 1, 3 and 5%). The moisture content ranged from 37.85% to 39.38%, exhibiting no significant differences between the groups. The pH level of VAE 5% added yanggaeng was the lowest, followed in order by 3% and 1% additions. Increasing the amount of VAE in yanggaeng tended to increase acidity. The lightness value of the Hunter color system decreased based on the amount of VAE concentrate added to yanggaeng. As VAE content increased, changes in hardness, gumminess, and brittleness were all significant (p<0.05). Characteristics of cohesiveness and springiness showed no significant differences. Total polyphenol content was the highest in VAE 5% added yanggaeng. Antioxidant activities such as DPPH radical scavenging activity of the control group was 2.29%, whereas groups with added VAE ranged from 7.52~28.05%. As VAE increased, antioxidative activity also increased. In the sensory evaluation, yanggaeng addition with VAE 3% had excellent scores for bitterness, moistness, and overall acceptability. Yanggaengs with moderate levels of VAE 3% are recommended (with respect to overall preference score) to take advantage of the functional properties of VAE without sacrificing consumer acceptability.

• Journal of Environmental Science International
• /
• v.28 no.5
• /
• pp.503-509
• /
• 2019

To prevent environmental pollution caused by leakage of leachate from waste landfill, vinyl acetate-ethylene (VAE) resin is applied to liner and cover materials to improve their performance. Styrene, styrene butadiene rubber, and VAE are widely used as polymer resins that have excellent water resistance and durability. Further, VAE resin is known to have additional advantages such as adhesion to nonpolar materials and resistance to saponification as a copolymer. In this study, the effect of VAE content on the properties of liner and cover materials was studied. The water and air content ratios, bending and compressive strengths, water absorption ratio, and coefficient of permeability of these materials were measured. The liner and cover materials with 4 wt% VAE showed good properties.

• Journal of Korea Technical Association of The Pulp and Paper Industry
• /
• v.45 no.5
• /
• pp.37-43
• /
• 2013

This study was focused on applying a new paper coating binder, vinyl acetate-ethylene(VAE) emulsion, with SB-latex and acrylic emulsion for paper coating application. VAE emulsion has a low monomer cost and is non-toxic chemical than conventional adhesive for paper coating such as styrene-butadiene latex( SB-latex) and acrylic emulsion. We conducted double coating in order to test VAE emulsion, which was applied on top surface only. The results showed that optical properties of the coated paper with VAE were similar with the SB-latex binders. In case of bonding strength, dry-pick of the coated paper with VAE showed almost same with other binders while wet-pick of the coated paper with VAE had a little bit lower strength than that with SB-latex.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.3
• /
• pp.351-358
• /
• 2022

Recently, Generative Adversarial Networks (GAN) and Variational AutoEncoders (VAE) have been applied to voice conversion that can make use of non-parallel training data. Especially, Conditional Cycle-Consistent Generative Adversarial Networks (CC-GAN) and Cycle-Consistent Variational AutoEncoders (CycleVAE) show promising results in many-to-many voice conversion among multiple speakers. However, the number of speakers has been relatively small in the conventional voice conversion studies using the CC-GANs and the CycleVAEs. In this paper, we extend the number of speakers to 100, and analyze the performances of the many-to-many voice conversion methods experimentally. It has been found through the experiments that the CC-GAN shows 4.5 % less Mel-Cepstral Distortion (MCD) for a small number of speakers, whereas the CycleVAE shows 12.7 % less MCD in a limited training time for a large number of speakers.

• Journal of Trauma and Injury
• /
• v.35 no.2
• /
• pp.76-83
• /
• 2022

Purpose: Ventilator-associated pneumonia is the most common nosocomial infection in patients with mechanical ventilation. In 2013, the new concept of ventilator-associated events (VAEs) replaced the traditional concept of ventilator-associated pneumonia. We analyzed risk factors for VAE occurrence and in-hospital mortality in trauma patients who received mechanical ventilatory support. Methods: In this retrospective review, the study population comprised patients admitted to the Jeju Regional Trauma Center from January 2020 to January 2021. Data on demographics, injury characteristics, and clinical findings were collected from medical records. The subjects were categorized into VAE and no-VAE groups according to the Centers for Disease Control and Prevention/National Healthcare Safety Network VAE criteria. We identified risk factors for VAE occurrence and in-hospital mortality. Results: Among 491 trauma patients admitted to the trauma center, 73 patients who received ventilator care were analyzed. Patients with a chest Abbreviated Injury Scale (AIS) score ≥3 had a 4.7-fold higher VAE rate (odds ratio [OR], 4.73; 95% confidence interval [CI], 1.46-17.9), and those with a glomerular filtration rate (GFR) <75 mL/min/1.73 m² had 4.1-fold higher odds of VAE occurrence (OR, 4.15; 95% CI, 1.32-14.1) and a nearly 4.2-fold higher risk for in-hospital mortality (OR, 4.19; 95% CI, 1.30-14.3). The median VAE-free duration of patients with chest AIS ≥3 was significantly shorter than that of patients with chest AIS <3 (P=0.013). Conclusions: Trauma patients with chest AIS ≥3 or GFR <75 mL/min/1.73 m² on admission should be intensively monitored to detect at-risk patients for VAEs and modify the care plan accordingly. VAEs should be closely monitored to identify infections early and to achieve desirable results. We should also actively consider modalities to shorten mechanical ventilation in patients with chest AIS ≥3 to reduce VAE occurrence.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.12
• /
• pp.1872-1879
• /
• 2022

Recently, with the development of artificial intelligence technology, research to use artificial intelligence to detect hacking attacks is being actively conducted. However, the fact that security data is a representative imbalanced data is recognized as a major obstacle in composing the learning data, which is the key to the development of artificial intelligence models. Therefore, in this paper, we propose a W-VAE oversampling technique that applies VAE, a deep learning generation model, to data extraction for oversampling, and sets the number of oversampling for each class through weight calculation using K-NN for sampling. In this paper, a total of five oversampling techniques such as ROS, SMOTE, and ADASYN were applied through NSL-KDD, an open network security dataset. The oversampling method proposed in this paper proved to be the most effective sampling method compared to the existing oversampling method through the F1-Score evaluation index.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.21 no.3
• /
• pp.27-34
• /
• 2017

Recently, the importance of the industrial warehouse floor has been increasing due to the development of the distribution and logistics industry. In this present study, an early-hardening polymer floor mortar which can compensate for the limitation of conventional cement based floor mortar regarding fluidity and long curing time was developed. In order to achieve the early-hardening of mortar characteristic ultra rapid hardening cement was used as binder. Four types of mixture proportions in accordance with the vinyl acetate ethylene(VAE) polymer contents with range from 10% to 20% and the other proto proportion without VAE polymer were designed. Mechanical experiments including the fluidity test, compressive strength test, bending test, bond test, and abrasion test were conducted for all mixture proportions. From the flow test result, it was possible to achieve the high flow with 250 mm by controlling the amount of superplasticizer. The incorporation of VAE polymer was found to affect the compressive strength reduction, however, the flexural strength was higher than that of the proto mixture, and it was evaluated to increase the compressive strength / flexural strength ratio. Moreover, at least 2.6 times higher bond strength and more than 4 times higher abrasion resistance were secured. From the mechanical experiments results, the optimum mixing ratio of the VAE polymer was determined to be 10%. As a result of application and monitoring, it shows that it has excellent resistance to cracking, discoloration, impact, and scratch as well as bond performance compared to the cement based floor mortar.

• Journal of the Institute of Convergence Signal Processing
• /
• v.23 no.4
• /
• pp.194-199
• /
• 2022

Preprocessing for high-quality data is required for high accuracy and usability in various and complex image data-based industries. However, when a contaminated hostile example that combines noise with existing image or video data is introduced, which can pose a great risk to the company, it is necessary to restore the previous damage to ensure the company's reliability, security, and complete results. As a countermeasure for this, restoration was previously performed using Defense-GAN, but there were disadvantages such as long learning time and low quality of the restoration. In order to improve this, this paper proposes a method using adversarial examples created through FGSM according to image segmentation in addition to using the VQ-VAE model. First, the generated examples are classified as a general classifier. Next, the unsegmented data is put into the pre-trained VQ-VAE model, restored, and then classified with a classifier. Finally, the data divided into quadrants is put into the 4-split-VQ-VAE model, the reconstructed fragments are combined, and then put into the classifier. Finally, after comparing the restored results and accuracy, the performance is analyzed according to the order of combining the two models according to whether or not they are split.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.6
• /
• pp.237-244
• /
• 2022

In this paper, we present VAE-based recommendation using online behavior log and purchase history to overcome data sparsity and cold start. To generate a variable for customers' purchase history, embedding and dimensionality reduction are applied to the customers' purchase history. Also, Variational Autoencoders are applied to online behavior and purchase history. A total number of 12 variables are used, and nDCG is chosen for performance evaluation. Our experimental results showed that the proposed VAE-based recommendation outperforms SVD-based recommendation. Also, the generated purchase history variable improves the recommendation performance.

• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.3
• /
• pp.94-102
• /
• 2023

Anomaly detection in human movements can improve safety in indoor workplaces. In this paper, we design a framework for detecting anomalous trajectories of humans in indoor spaces based on a variational autoencoder (VAE) with Bi-LSTM layers. First, the VAE is trained to capture the latent representation of normal trajectories. Then the abnormality of a new trajectory is checked using the trained VAE. In this step, the anomaly score of the trajectory is determined using the trajectory reconstruction error through the VAE. If the anomaly score exceeds a threshold, the trajectory is detected as an anomaly. To select the anomaly threshold, a new metric called D-score is proposed, which measures the difference between recall and precision. The anomaly threshold is selected according to the minimum value of the D-score on the validation set. The MIT Badge dataset, which is a real trajectory dataset of workers in indoor space, is used to evaluate the proposed framework. The experiment results show that our framework effectively identifies abnormal trajectories with 81.22% in terms of the F1-score.