• Clean Technology
• /
• v.25 no.3
• /
• pp.256-262
• /
• 2019

Hydroxylammonium nitrate (HAN)-based liquid propellants are attracting attention as environmentally friendly propellants because they are not carcinogens and the combustion gases have little toxicity. The catalyst used to decompose the HAN-based liquid propellant in a thruster must have both low temperature activity and high heat resistance. The objective of this study is to prepare an Ru/alumina/metal foam catalyst by supporting alumina slurry on the surface of NiCrAl metal foam using a washing coating method and then to support a ruthenium precursor thereon. The decomposition activity of a HAN aqueous solution of the Ru/alumina/metal foam catalyst was evaluated. The effect of the number of repetitive coatings of alumina slurry on the physical properties of the alumina/metal foam was analyzed. As the number of alumina wash coatings increased, mesopores with a diameter of about 7 nm were well-developed, thereby increasing the surface area and pore volume. It was optimal to repeat the wash coating alumina on the metal foam 12 times to maximize the surface area and pore volume of the alumina/metal foam. Mesopores were also well developed on the surface of the Ru/alumina/metal foam catalyst. It was found that the metal form itself without the active metal and alumina can promote the decomposition reaction of the HAN aqueous solution. In the case of the Ru/alumina/metal foam-550 catalyst, the decomposition onset temperature was significantly lowered compared with that of the thermal decomposition reaction, and ${\Delta}P$ could be greatly increased in the decomposition of the HAN aqueous solution. However, when the catalyst was calcined at $1,200^{\circ}C$, the catalytic activity was lowered inevitably because the surface area and pore volume of the catalyst were drastically reduced and Ru was sintered. Further research is needed to improve the heat resistance of Ru/alumina/metal foam catalysts.

• Journal of Mushroom
• /
• v.17 no.3
• /
• pp.119-125
• /
• 2019

This study aims to report the results for the analysis of the growth environment by applying smart farm technology to "Chunchu No 2" farmers in order to develop an optimal growth model for precision cultivation of bottle-grown oyster mushrooms. The temperature, humidity, carbon dioxide concentration, and illumination data were collected and analyzed using an environmental sensor installed to obtain growth environment data from the oyster mushroom cultivator. Analysis of the collected temperature data revealed that the temperature at the time of granulation was $19.5^{\circ}C$ after scraping, and the mushroom was generated and maintained at about $21^{\circ}C$ until the bottle was flipped. When the fruiting body grew and approached harvest time, mushrooms were harvested while maintaining the temperature between $14^{\circ}C$ and $18^{\circ}C$. The humidity was maintained at almost 100% during the complete growth stage. Carbon dioxide concentration gradually increased until 3 days after the beginning of cultivation, and then increased rapidly to almost 5,500 ppm. From the 6th day, carbon dioxide concentration was gradually decreased through ventilation and was maintained at 1,600 ppm during harvest. Light intensity of 8 lux was irradiated up to day 6 after seeding, and growth was then continued while periodically irradiating 4 lux light. The fruiting body characteristics of "Chunchu No 2" cultivated in the farmhouse were as follows: pileus diameter of 26.5 mm and thickness of 4.9 mm, stipe thickness of 8.9 mm, and length of 68.7 mm. The fruiting body yield was 166.8 g/850 ml, and the individual weight was 12.8 g/10 units.

• Journal of Intelligence and Information Systems
• /
• v.25 no.3
• /
• pp.239-251
• /
• 2019

Statistical methods such as moving averages, Kalman filtering, exponential smoothing, regression analysis, and ARIMA (autoregressive integrated moving average) have been used for stock market predictions. However, these statistical methods have not produced superior performances. In recent years, machine learning techniques have been widely used in stock market predictions, including artificial neural network, SVM, and genetic algorithm. In particular, a case-based reasoning method, known as k-nearest neighbor is also widely used for stock price prediction. Case based reasoning retrieves several similar cases from previous cases when a new problem occurs, and combines the class labels of similar cases to create a classification for the new problem. However, case based reasoning has some problems. First, case based reasoning has a tendency to search for a fixed number of neighbors in the observation space and always selects the same number of neighbors rather than the best similar neighbors for the target case. So, case based reasoning may have to take into account more cases even when there are fewer cases applicable depending on the subject. Second, case based reasoning may select neighbors that are far away from the target case. Thus, case based reasoning does not guarantee an optimal pseudo-neighborhood for various target cases, and the predictability can be degraded due to a deviation from the desired similar neighbor. This paper examines how the size of learning data affects stock price predictability through k-nearest neighbor and compares the predictability of k-nearest neighbor with the random walk model according to the size of the learning data and the number of neighbors. In this study, Samsung electronics stock prices were predicted by dividing the learning dataset into two types. For the prediction of next day's closing price, we used four variables: opening value, daily high, daily low, and daily close. In the first experiment, data from January 1, 2000 to December 31, 2017 were used for the learning process. In the second experiment, data from January 1, 2015 to December 31, 2017 were used for the learning process. The test data is from January 1, 2018 to August 31, 2018 for both experiments. We compared the performance of k-NN with the random walk model using the two learning dataset. The mean absolute percentage error (MAPE) was 1.3497 for the random walk model and 1.3570 for the k-NN for the first experiment when the learning data was small. However, the mean absolute percentage error (MAPE) for the random walk model was 1.3497 and the k-NN was 1.2928 for the second experiment when the learning data was large. These results show that the prediction power when more learning data are used is higher than when less learning data are used. Also, this paper shows that k-NN generally produces a better predictive power than random walk model for larger learning datasets and does not when the learning dataset is relatively small. Future studies need to consider macroeconomic variables related to stock price forecasting including opening price, low price, high price, and closing price. Also, to produce better results, it is recommended that the k-nearest neighbor needs to find nearest neighbors using the second step filtering method considering fundamental economic variables as well as a sufficient amount of learning data.

• Journal of the Korean Applied Science and Technology
• /
• v.36 no.3
• /
• pp.853-865
• /
• 2019

In this study, the skin permeability was measured by adding skin penetrating peptides, arginine oligomers R4(tetra-D-arginine), R6(hexa-D-arginine) to little skin-permeable wrinkle improvement peptides GHK, GHK-Cu, and Pal-GHK liposomes, and the results were analyzed by the following six cases. (1) In cases where only wrinkle improvement peptides GHK, GHK-Cu, and Pal-GHK were contained liposomes; the final cumulative permeations in 24 hours were 6.05%, 7.4%, and 8.83% respectively. (2) In cases where arginine oligomers R4, R6 were added to GHK liposomes; the final cumulative permeations in 24 hours were 13.63% and 7.68%. (3) In cases where R4, R6 were added to GHK-Cu liposomes; the final cumulative permeations in 24 hours were 15.46% and 8.64%. (4) In cases where R4, R6 were added to Pal-GHK liposomes; the final cumulative permeations in 24 hours were 16.9% and 10.67%. (5) In cases where R4 were added to GHK, GHK-Cu, and Pal-GHK liposomes; the final cumulative permeations in 24 hours were 13.63%, 15.46%, and 16.9% respectively. (6) In cases where R6 were added to GHK, GHK-Cu, and Pal-GHK liposomes; the final cumulative permeations in 24 hours were 7.68%, 8.64%, and 10.67% respectively. This experiment showed that skin absorption of GHK was increased by copper ion (Cu2+) and palmitic acid and skin absorption of wrinkle improvement peptides was enhanced by cell penetrating peptides, and R4 showed higher effect than R6 in GHK, GHK-Cu and Pal-GHK. Through this process, we propose broad use and application in wrinkle improvement functional cosmetics by presenting the optimal conditions for increasing skin absorption of GHK, GHK-Cu, thus maximizing its efficacy.

• Journal of Plant Biotechnology
• /
• v.46 no.1
• /
• pp.22-31
• /
• 2019

The objective of this study was to investigate the suitable parts for callus induction and optimal concentrations of growth regulators contained in the medium affecting shooting and rooting Echeveria laui and Echeveria elegans for in vitro mass production. To determine the suitable plant parts for callus induction, the leaves were divided into upper, medium and bottom parts and cultured on MS medium at different concentrations with $0{\sim}2mgL^{-1}\;NAA$ and $0{\sim}4 mgL^{-1}BA$. The upper and middle parts of leaves both showed 100% callus formation rate with $NAA\;1\;mgL^{-1}$ and $BA\;1\;mgL^{-1}$ treatment in E. laui. The middle parts of leaves showed 83.3% callus formation rate at $NAA\;2\;mgL^{-1}$ and BA 4 mgL-1 treatment in E. elegans. The shoot induction rate from callus was highest at $NAA\;0.1\;mgL^{-1}$ and $BA\;3\;mgL^{-1}$ treatment in E. laui and $NAA\;0.3\;mgL^{-1}$ in E. elegans. In addition, the number of shoots formation was 10.4 shoots high in $NAA\;1\;mgL^{-1}$ and $BA\;1\;mgL^{-1}$ treatment in E. laui and 12.0 shoots in most effective $NAA\;1\;mgL^{-1}$ and $BA\;0.1\;mgL^{-1}$ treatment in E. elegans. In the case of acclimatization of regenerated plant, growth characteristics did not show any significant difference (35 ~ 55%) shading with respect to the different ratio of substrate mixture, and it was determined that would be appropriate considered plant height and appearance preference of E. laui and E. elegans. It was established that the optimization of culture condition was responsible for the mass propagation in vitro cultures of E. laui and E. elegans.

• Journal of Life Science
• /
• v.29 no.1
• /
• pp.76-83
• /
• 2019

In this study, the optimal growth and poly(3-hydroxybutyrate) (PHB) biosynthesis of Pseudomonas sp. EML2 were established using waste frying oil (WFO) as a cheap carbon source. The fatty acid composition of WFO and fresh frying oil (FFO) were analyzed by gas chromatography. The unsaturated and saturated fatty acid contents of the FFO were 82.6% and 14.9%, respectively. These contents changed in the WFO. The compositional change in the unsaturated fatty acid content in the WFO was due to a change in its chemical and physical properties resulting from heating, an oxidation reaction, and hydrolysis. The maximum dry cell weight (DCW) and PHB yield (g/l) of the isolated strain Pseudomonas sp. EML2 were confirmed under the following culture conditions: 30 g/l of WFO, 0.5 gl of $NH_4Cl$, pH 7, and $20^{\circ}C$. Based on this, the growth and PHB yield of Pseudomonas sp. EML2 were confirmed by 3 l jar fermentation. After the cells were cultured in 30 g/l of WFO for 96 h, the DCW, PHB content, and PHB yield of Pseudomonas sp. EML2 were 3.6 g/l, 73 wt%, and 2.6 g/l, respectively. Similar results were obtained using 30 g/l of FFO as a carbon source control. Using the FFO, the DCW, PHB content, and PHB yield were 3.4 g/l, 70 wt%, and 2.4 g/l, respectively. Pseudomonas sp. EML2 and WFO may be a new candidate and substrate, respectively, for industrial production of PHB.

• Journal of Mushroom
• /
• v.17 no.2
• /
• pp.52-63
• /
• 2019

Agrocybe spp. belongs to the Agaricales order, Bolbitiaceae family, and Agrocybe genus. In Korea, so far, it has been cultivated through bottle cultivation; therefore, this study was conducted for the development of a new cultivar using the bag cultivation method for quantitative improvement. Thirty-three gene resources of Agrocybe spp. were collected and their quantity and characteristics of bag cultivation were examined. Next, 5 kinds of crossing parents were selected based on the cultivation period and shape of the fruit body. Seven strains were selected through 3 cross combinations. The 7 selected strains and the comparative cultivar 'Cham' were cultivated in a bag. As a result, the cultivation period was 49 days for 'JBAC15-1' and 50 days for 'JBAC15-6' which are 4 days and 3 days less than the cultivation period of the comparative cultivar 'Cham'(53 days), respectively. Cultivation periods of other strains except for 'JBAC15-1' and 'JBAC15-1' were longer than that of the comparative cultivar'Cham'. The best ratio of primordia formation among the selected strains was found to be that of 'JBAC15-1' with 96.1% followed by 'JBAC15-6' with 94.5%. These rates were 3.1% and 1.5% higher than the ratio of primordia formation of the comparative cultivar 'Cham', which is found to be 93.0%. The quantity was maximum in the 'JBAC15-1' cultivar with 176.8 g per bag followed by 'JBAC15-6' with 168.7 g per bag. The quantities were 10% and 5% more than the comparative cultivar 'Cham' with 160.7 g per bag. Based on these results, 'JBAC15-1' and 'JBAC15-6', which had shorter cultivation periods and more quantities than the comparative cultivar 'Cham' were finally selected. For the selected strains of 'JBAC15-1' and 'JBAC15-6', mycelial growth was observed to be optimal on PDA medium and the optimum temperature was $27.5^{\circ}C$. The optimum pH was pH 5 for 'JBAC15-1' and pH 6 for 'JBAC15-6'. The color of the pileus of the fruit body was dark brown in 'JBAC15-1' and 'Cham' and light brown in 'JBAC15-6'. The pileus was hemispherical in shape in both 'JBAC15-1' and 'Cham'. However, the colors of the stem were different - light brown in 'JBAC15-1', white in 'JBAC15-6', and ivory in the comparative cultivar 'Cham'.

• Food Engineering Progress
• /
• v.21 no.1
• /
• pp.42-48
• /
• 2017

The purpose of this study was to verify the drying characteristics of steamed sweet potato and to establish optimal drying conditions for far-infrared drying of steamed sweet potato. 4 kg of steamed sweet potato was sliced to thicknesses of 8 and 10 mm, and dried by a far-infrared dryer until a final moisture content of $25{\pm}0.5%$. The far-infrared dryer conditions were an air velocity of 0.6, 0.8 m/s and drying temperature of 60, 70, and $80^{\circ}C$. The results can be summarized as follows. The drying time tended to be reduced as temperature and air velocity for drying increased. The Lewis and Modified Wang and Singh models were found to be suitable for drying of steamed sweet potato by a far-infrared dryer. The color difference was 35.09 on the following conditions: Thickness of 8 mm, temperature of $80^{\circ}C$, and air velocity of 0.8 m/s. The highest sugar content ($59.11^{\circ}Brix$) was observed on the conditions of a thickness of 8 mm, temperature of 80, and air velocity of 0.8 m/s. Energy consumption decreased on the conditions of higher temperature, slower air velocity, and thinner steamed sweet potato.

• Korean Journal of Food Science and Technology
• /
• v.53 no.2
• /
• pp.195-203
• /
• 2021

Aureobasidium pullulans, a black yeast, produces pullulan, a linear α-glucan composed of maltotriose repeating units linked by α(1→6)-glycosidic linkages. Pullulan can be widely used in food, cosmetic, and biotechnology industries. In this study, we isolated eight strains of A. pullulans from Forsythia koreana, Magnolia kobus DC., Spiraea prunifolia var. simpliciflora, Cornus officinalis, Cerasus, and Hippophae rhamnoides. Among them, A. pullulans MR was selected as the best pullulan producer. The effects of a carbon source, a nitrogen source, and pH on pullulan production were examined. The optimal cultivation conditions for pullulan production by A. pullulans MR were determined by response surface methodology as 15% sucrose, 0.4% soy peptone, and an initial pH of 7 at 26℃. Under these conditions, the predicted pullulan production was 47.6 g/L, which was very close to the experimental data (48.9 g/L).

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.157-173
• /
• 2021

With the development of information technology, the amount of available information increases daily. However, having access to so much information makes it difficult for users to easily find the information they seek. Users want a visualized system that reduces information retrieval and learning time, saving them from personally reading and judging all available information. As a result, recommendation systems are an increasingly important technologies that are essential to the business. Collaborative filtering is used in various fields with excellent performance because recommendations are made based on similar user interests and preferences. However, limitations do exist. Sparsity occurs when user-item preference information is insufficient, and is the main limitation of collaborative filtering. The evaluation value of the user item matrix may be distorted by the data depending on the popularity of the product, or there may be new users who have not yet evaluated the value. The lack of historical data to identify consumer preferences is referred to as data sparsity, and various methods have been studied to address these problems. However, most attempts to solve the sparsity problem are not optimal because they can only be applied when additional data such as users' personal information, social networks, or characteristics of items are included. Another problem is that real-world score data are mostly biased to high scores, resulting in severe imbalances. One cause of this imbalance distribution is the purchasing bias, in which only users with high product ratings purchase products, so those with low ratings are less likely to purchase products and thus do not leave negative product reviews. Due to these characteristics, unlike most users' actual preferences, reviews by users who purchase products are more likely to be positive. Therefore, the actual rating data is over-learned in many classes with high incidence due to its biased characteristics, distorting the market. Applying collaborative filtering to these imbalanced data leads to poor recommendation performance due to excessive learning of biased classes. Traditional oversampling techniques to address this problem are likely to cause overfitting because they repeat the same data, which acts as noise in learning, reducing recommendation performance. In addition, pre-processing methods for most existing data imbalance problems are designed and used for binary classes. Binary class imbalance techniques are difficult to apply to multi-class problems because they cannot model multi-class problems, such as objects at cross-class boundaries or objects overlapping multiple classes. To solve this problem, research has been conducted to convert and apply multi-class problems to binary class problems. However, simplification of multi-class problems can cause potential classification errors when combined with the results of classifiers learned from other sub-problems, resulting in loss of important information about relationships beyond the selected items. Therefore, it is necessary to develop more effective methods to address multi-class imbalance problems. We propose a collaborative filtering model using CGAN to generate realistic virtual data to populate the empty user-item matrix. Conditional vector y identify distributions for minority classes and generate data reflecting their characteristics. Collaborative filtering then maximizes the performance of the recommendation system via hyperparameter tuning. This process should improve the accuracy of the model by addressing the sparsity problem of collaborative filtering implementations while mitigating data imbalances arising from real data. Our model has superior recommendation performance over existing oversampling techniques and existing real-world data with data sparsity. SMOTE, Borderline SMOTE, SVM-SMOTE, ADASYN, and GAN were used as comparative models and we demonstrate the highest prediction accuracy on the RMSE and MAE evaluation scales. Through this study, oversampling based on deep learning will be able to further refine the performance of recommendation systems using actual data and be used to build business recommendation systems.