• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.23-45
• /
• 2020

Big data is creating in a wide variety of fields such as medical care, manufacturing, logistics, sales site, SNS, and the dataset characteristics are also diverse. In order to secure the competitiveness of companies, it is necessary to improve decision-making capacity using a classification algorithm. However, most of them do not have sufficient knowledge on what kind of classification algorithm is appropriate for a specific problem area. In other words, determining which classification algorithm is appropriate depending on the characteristics of the dataset was has been a task that required expertise and effort. This is because the relationship between the characteristics of datasets (called meta-features) and the performance of classification algorithms has not been fully understood. Moreover, there has been little research on meta-features reflecting the characteristics of multi-class. Therefore, the purpose of this study is to empirically analyze whether meta-features of multi-class datasets have a significant effect on the performance of classification algorithms. In this study, meta-features of multi-class datasets were identified into two factors, (the data structure and the data complexity,) and seven representative meta-features were selected. Among those, we included the Herfindahl-Hirschman Index (HHI), originally a market concentration measurement index, in the meta-features to replace IR(Imbalanced Ratio). Also, we developed a new index called Reverse ReLU Silhouette Score into the meta-feature set. Among the UCI Machine Learning Repository data, six representative datasets (Balance Scale, PageBlocks, Car Evaluation, User Knowledge-Modeling, Wine Quality(red), Contraceptive Method Choice) were selected. The class of each dataset was classified by using the classification algorithms (KNN, Logistic Regression, Nave Bayes, Random Forest, and SVM) selected in the study. For each dataset, we applied 10-fold cross validation method. 10% to 100% oversampling method is applied for each fold and meta-features of the dataset is measured. The meta-features selected are HHI, Number of Classes, Number of Features, Entropy, Reverse ReLU Silhouette Score, Nonlinearity of Linear Classifier, Hub Score. F1-score was selected as the dependent variable. As a result, the results of this study showed that the six meta-features including Reverse ReLU Silhouette Score and HHI proposed in this study have a significant effect on the classification performance. (1) The meta-features HHI proposed in this study was significant in the classification performance. (2) The number of variables has a significant effect on the classification performance, unlike the number of classes, but it has a positive effect. (3) The number of classes has a negative effect on the performance of classification. (4) Entropy has a significant effect on the performance of classification. (5) The Reverse ReLU Silhouette Score also significantly affects the classification performance at a significant level of 0.01. (6) The nonlinearity of linear classifiers has a significant negative effect on classification performance. In addition, the results of the analysis by the classification algorithms were also consistent. In the regression analysis by classification algorithm, Naïve Bayes algorithm does not have a significant effect on the number of variables unlike other classification algorithms. This study has two theoretical contributions: (1) two new meta-features (HHI, Reverse ReLU Silhouette score) was proved to be significant. (2) The effects of data characteristics on the performance of classification were investigated using meta-features. The practical contribution points (1) can be utilized in the development of classification algorithm recommendation system according to the characteristics of datasets. (2) Many data scientists are often testing by adjusting the parameters of the algorithm to find the optimal algorithm for the situation because the characteristics of the data are different. In this process, excessive waste of resources occurs due to hardware, cost, time, and manpower. This study is expected to be useful for machine learning, data mining researchers, practitioners, and machine learning-based system developers. The composition of this study consists of introduction, related research, research model, experiment, conclusion and discussion.

• Korean Chemical Engineering Research
• /
• v.53 no.1
• /
• pp.31-38
• /
• 2015

At present, carbon dioxide ($CO_2$) emission, which causes global warming, is a major issue all over the world. To reduce $CO_2$ emission directly, commercial deployment of $CO_2$ separation processes has been attempted in industrial plants, such as power plant, oil refinery and steelmaking plant. Besides, several studies have been done on indirect reduction of $CO_2$ emission from recycle of reducing gas (carbon monoxide or hydrogen containing gas) in the plants. Unlike many competing gas separation technologies, pressure swing adsorption (PSA) and membrane filtration are commercially used together or individually to separate a single component from the gas mixture. However, there are few studies on operation of sequential separation process of multi-component gas which has more than two target gas products. In this paper, process simulation model is first developed for two available configurations: $CO_2$ PSA-CO PSA-$H_2$ PSA and $CO_2$ PSA-CO PSA-$H_2$ membrane. Operation optimization and economic evaluation of the processes are also performed. As a result, feed gas contains about 14% of $H_2$ should be used as fuel than separating $H_2$, and $CO_2$ separation should be separated earlier than CO separation when feed gas contains about 30% of $CO_2$ and CO. The simulation results can help us to find an optimal process configuration and operation condition for separation of multicomponent gas with $CO_2$, CO, $H_2$ and other gases.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.39 no.6
• /
• pp.98-109
• /
• 2011

Habitat restoration initiative is an important factor for the on-site preservation of butterflies. A new habitat necessitates the investigation of the ecology of the butterfly, the landscape of the habitat, the work process, and a cost analysis. In this study, these factors were analyzed through1) re-consideration of the secured results from three years of field work, and 2) a habitat area estimation conducted according to the presence or absence of larvae and adults via a Parnassius bremeri Bremer survey. The investigation of the natural habitat, done for both larvae and adult butterflies found in South Korea, suggested that multiple patches with a minimum size exceeding $300m^2$ and with an average size of about $1,600m^2$were required for survival of this butterfly. Therefore, more than five patches should be considered for butterfly habitat and patches should have similarity to the present habitat environment with transplantation plans for diet plants and honeydew plants. In addition, to activate the migration of the butterfly, cutting down trees that are obstacles to migration, minimizing the distance between patches and the addition of a stepping patch are also required. A patch connectivity analysis showed that patches should be located within 300m of neighboring patches, as the migration of the butterfly will fail if it is more than 600m. Additionally, more than $10,000m^2$ for a single patch or a patch network composed more than five patches over $1,600m^2$ in size were recommended, with a total patch size of $8,000m^2$. Additionally, neighboring multiple patch networks are more desirable than a single habitat in the event a cyclic habitat pattern arises. In this study, we suggest a habitat restoration and optimal prerequisites for a butterfly habitat. It is expected that this research will lead to the creation of a good model for the restoration of butterfly species.

• Journal of Bio-Environment Control
• /
• v.23 no.2
• /
• pp.109-115
• /
• 2014

This study examined the uplift bearing capacity of spiral steel pegs according to the degree of soil compaction and embedded depth in a small-scaled lab test. As a result, their uplift bearing capacity increased according to the degree of soil compaction and embedded depth. The uplift bearing capacity under the ground condition of 85% compaction rate especially recorded 48.9 kgf, 57.9 kgf, 86.2 kgf and 116.6 kgf at embedded depth of 25 cm, 30 cm, 35 cm and 40 cm, respectively, being considerably higher than under other ground conditions. There were huge differences in the uplift bearing capacity of spiral steel pegs according to the compaction conditions of ground. Their maximum uplift bearing capacity was 116.6 kgf under the ground condition of 85% compaction rate and at embedded depth of 40 cm, and it is very high considering the data of spiral steel pegs. It is thus estimated that wind damage can be effectively reduced by careful maintenance of ground condition surrounding spiral steel pegs. In addition, spiral steel pegs will be able to make a contribution to greenhouse structural stability if proper installation methods are provided including the number and interval according to the types of greenhouse as well as fixation of plastic film. The findings of the study indicate that the optimal effects of spiral steel pegs for greenhouse can be achieved at embedded depth of more than 35cm and compaction degree of more than 85%. The relative density of the model ground in the test was 67% at compaction rate of 85%.

• Clinical and Experimental Reproductive Medicine
• /
• v.26 no.2
• /
• pp.137-148
• /
• 1999

Purpose of the present study was to find the optimal ovulation induction medicine for the maturation and development of immature oocytes and culture media for 2-cell embryos in the mouse model. ICR female mouse aged 6 to 8 weeks, were stimulated with 5 IU PMSG injection. At 47 to 50 hour post-PMSG injection, ovaries were dissected out and oocytes-cumulus complexes were punctured. The oocyte-cumulus complexes were cultured in media containing various ovulation induction medicine, CC, HMG and Metrodin for 18 hours. Female ICR mice were stimulated with 5 IU PMSG and 48 hours later were injected 5 IU of hCG, then female and male mice were mated. At 48 hour post-hCG injection, oviducts were dissected out and 2-cell embryos were flushed. The 2-cell embryos were cultured in various media, Ham's F-10 media of milli-Q water $(3^{\circ})$, Ham's F-10 media of HPLC (high performance liquid chromatography, Baxter) water, Medicult media, HTF (human tubal fluid) media for 96 hours. The results were as follows. 1. When the oocytes-cumulus complexes were cultured in $10^{-9}{\mu}g/ml{\sim}10^{-8}{\mu}g/ml$ of CC, those were suppressed in meiotic maturation $(28.2{\sim}33.7%)$. Whereas the oocytes-cumulus complexes were cultured in $10^{-7}{\mu}g/ml{\sim}10^{-4}{\mu}g/ml$, these were not effected in meiotic maturation $(54.5{\sim}72.7%)$. 2. When the oocytes-cumulus complexes were cultured in $10^{-4}{\mu}g/ml{\sim}10^{-1}{\mu}g/ml$ of Metrodin, those were suppressed in meiotic maturation $(35.7{\sim}41.5%)$. Meanwhile the oocytes-cumulus complexes were cultured in $10^{-7}{\mu}g/ml{\sim}10^{-5}{\mu}g/ml$, those were not effected in meiotic maturation $(54.2{\sim}70.3%)$. 3. When the oocytes-cumulus complexes were cultured in $10^{-5}{\mu}g/ml{\sim}10^{-4}{\mu}g/ml$ of HMG, those were suppressed in meiotic maturation $(48.2{\sim}50.4%)$. As being cultured in $10^{-7}{\mu}g/ml{\sim}10^{-6}{\mu}g/ml$, increased in meiotic maturation $(75.8{\sim}80.7%)$. 4. When the 2-cell embryos were cultured in Ham's F-10 media of milli-Q water $(3^{\circ})$, Ham's F-10 media of HPLC (high performance liquid chromatograpy, Baxter) water, Medicult media, HTF (human tubal fluid) media, developmental rates to blastocyst and hatching for 96 hour were 50.0%, 45.2%, 71.5% and 95.6%, respectively.

• Korean Security Journal
• /
• no.11
• /
• pp.203-226
• /
• 2006

A highly sophisticated expertise and systematic and integrated management of security operation are essential for a crowded stadium. a special object of security guarding. Nonetheless, the recent incident in a singing concert hall reveals the overall problem like the lacking safety management system. lacking deployment of professional security personnel, absence of safety manuals and safety measures, as well as the lack of professionalism of private sector security companies. In this study, we presented three categories that needed improvement, like the legal and institutional improvement, improvement of policy and improvement of operation which are required to set up the model to operate the optimal private sector security duties. For the revision of law and institution for a better and more desirable method, we discussed the revision of related laws and regulations pursuant to the security operation at places where events are held, including the revision of law on security guarding work, regulation on common housing management, uniformity of security guarding, and law on performance. For the improvement of policies, we discussed the introduction of security consultants, strengthening the security instructor system, expansion of relevant organizations, establishment of mutual cooperation, privatization of profitable events, improvement of awareness about the security activities provided by private sector, policy for the professionalism of private security operation, expansion of security exhibition and seminar. For the improvement of operation. we discussed professional security techniques. such as the technique of security consulting, the application of CPTED technique, the technique for the integration of system, the method of operation, the establishment of a system to support public security operations and volunteers, establishment of a manual for security guarding performance, modernization and high tech-oriented equipment, organization of security guarding entity in which the industry, academic society and government participates together.

• KSBB Journal
• /
• v.14 no.3
• /
• pp.328-336
• /
• 1999

A green microalga, Chlorella vulgaris UTX 259, was cultivated in a bench-scale raceway pond. During the culture, 15%(v/v) $CO_2$ was supplied and industrial wastewater discharged from a steel-making plant was used as a culture medium. In a small scale culture bottle, the microalga grew up to 1.8 g $dm^{-3}$ of cell concentration and ammonia was completely removed from the wastewater with an yield coefficient of 25.7 g dry cell weight $g^{-1}\;NH_3-N$. During the bottle-culture, microalga was dominant over heterotrophic microorganisms in the culture medium. Therefore, the amount of carbon dioxide fixation could be estimated from the change of dry cell weight. In a semi-continuous operation of raceway pond with intermittent lighting (12 h light and 12 h dark), increase of dilution rate resulted in increase of the ammonia removal rate as well as the $CO_2$ fixation rate but the ammonia removal efficiency decreased. Ammonia was not completely removed from the medium (wastewater) of raceway pond which was operated in a batch mode under a light intensity up to 20 klux. The incomplete removal of ammonia was believed due to insufficient light supply. A mathematical model, capable of predicting experimental data, was developed in order to simulate the performance of the raceway pond under the light intensity of sun during a bright daytime. Simulation results showed that the rates of $CO_2$ fixation and ammonia removal could be enhanced by increasing light intensity. According to the simulation, 80 mg $dm^{-3}$ of ammonia in the medium could be completely removed if the light intensity was over 60 klux with a continuous lighting. Under the optimal operating condition determined by the simulation, the rates of carbon dioxide fixation and ammonia removal in the outdoor operation of raceway pond were estimated as high as $24.7 g m^{-2} day^{-1}$ and $0.52 g NH_3-N m^{-2} day^{-1}$, respectively.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.45 no.9
• /
• pp.1358-1365
• /
• 2016

The purpose of this study was to verify the drying characteristics of seasoned red pepper sauce and establish optimal drying conditions for far infrared drying of seasoned red pepper sauce. Seasoned red pepper sauce, which was dried by heated air, was used. One kg of seasoned red pepper was spread at thicknesses of 10 and 20 mm and dried by a far infrared dryer until a final moisture content of $15{\pm}0.5%$. The far infrared dryer conditions were air velocity of 0.6, 0.8 m/s and drying temperatures of 60, 70, and $80^{\circ}C$. The drying models were estimated using a determination coefficient and root mean square error. Drying characteristics were analyzed based on factors such as drying rate, color changes, content of capsaicinoids, and energy consumption. The results can be summarized as follows. The drying rate (that is, drying time) tended to be reduced as temperature and air velocity for drying increased. The Page and Henderson models were suitable for drying of seasoned red pepper sauce by a far infrared dryer. Redness decreased after far infrared drying under all experimental conditions. The color difference was 18.18 under the following conditions: thickness 20 mm, temperature $70^{\circ}C$, and air velocity 0.8 m/s. This value was slightly higher than those under other far infrared drying conditions. The capsaicinoid properties of seasoned red pepper sauce decreased under all far infrared drying conditions. The highest capsaicin (19.91 mg/100 g) and dihydrocapsaicin (12.87 mg/100 g) contents were observed at a thickness of 10 mm, temperature of $80^{\circ}C$, and air velocity of 0.8 m/s. Energy consumption decreased with higher temperature, slower air velocity, and thinner seasoned red pepper sauce.

• Journal of Mushroom
• /
• v.16 no.3
• /
• pp.155-161
• /
• 2018

In this study, we analyze the growth environment using smart farm technology in order to develop the optimal growth model for the precision cultivation of the bottle-grown oyster mushroom 'Suhan'. Experimental farmers used $88m^2$ of bed area, 2 rows and 5 columns of shelf shape, 5 hp refrigerator, 100T of sandwich panel for insulation, 2 ultrasonic humidifiers, 12 kW of heating, and 5,000 bottles for cultivation. Data on parameters such as temperature, humidity, carbon dioxide concentration, and illumination, which directly affect mushroom growth, were collected from the environmental sensor part installed at the oyster mushroom cultivator and analyzed. It was found that the initial temperature at the time of granulation was $22^{\circ}C$ after the scraping, and the mushroom was produced and maintained at about $25^{\circ}C$ until the bottle was flipped. On fruiting body formation, mushrooms were harvested while maintaining the temperature between $13^{\circ}C$ and $15^{\circ}C$. Humidity was approximately 100% throughout the growth stage. Carbon dioxide concentration gradually increased until 3 days after the beginning of cultivation, and then increased rapidly to approximately 2,600 ppm. From the 6th day, $CO_2$ concentration was gradually decreased through ventilation and maintained at 1,000 ppm during the harvest. Light was not provided at the initial stage of oyster mushroom cultivation. On the $3^{rd}$ and $4^{th}$ day, mushrooms were irradiated by 17 lux light. Subsequently, the light intensity was increased to 115-120 lux as the growth progressed. Fruiting body characteristics of 'Suhan' cultivated in a farmhouse were as follows: Pileus diameter was 30.9 mm and thickness was 4.5 mm; stipe thickness was 11.0 mm and length was 76.0 mm; stipe and pileus hardness was 0.8 g/mm and 2.8 g/mm, respectively; L values of the stipe and pileus were 79.9 and 52.3, respectively. The fruiting body yield was 160.2 g/850 ml, and the individual weight was 12.8 g/10 unit.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.44 no.5
• /
• pp.59-67
• /
• 2016

In landscape design by public institutions, although the costs and species of landscape trees stipulated by the Korean Public Procurement Service(PPS) are generally adhered to, the PPS regulations about planting trees with well-developed rootlets are almost entirely neglected. This study aimed to evaluate the performance of buried container modules, which are a new technology and product in landscape production that is able to reduce the defect rate while complying with regulations. To this end, this study measured rootlet density, rootlet development length, rootlet survival rate on excavation, and impairments of tree growth for 3 months after root pruning, and compared these variables for the container modules with those for trees that underwent root pruning in bare ground, and those that were cultivated in a container above ground. The results were as follows: First, the rootlet density was 88% for the trees in container modules, which was very high. Trees that underwent standard root pruning in bare ground had a somewhat lower density of 64%. Meanwhile, the trees that were cultivated in pots above ground died, invalidating measurement. Second, in terms of rootlet development and rootlet survival rate, the trees in container modules showed a mean length of 10.4cm, and 100% survival rate, indicating that there was no rootlet damage caused by excavation. For the trees that only underwent root pruning in bare ground, the mean length was 25.6cm and the rootlet survival rate was only half that of the trees in container modules, at 56%, demonstrating considerable damage. Rootlet development did not occur at all in the trees grown in pots. Third, the trees in container modules and those that underwent root pruning in bare ground did not show any deaths during the root pruning process, or any impairments such as stunted leaf growth. Conversely, the trees grown in pots nearly all died, and severe impairments of tree growth were observed. As shown by the results above, when we evaluated the performance of buried container modules, they showed the most outstanding performance of the three models tested in this study. The container modules prevent defects by stimulating early rooting in environments that with poor conditions for growth, or in trees that are not suited to the summer environment Therefore, it is expected that they would be an optimal means by which to enable compliance with rules such as the regulation presented by the PPS.