• Journal of Intelligence and Information Systems
• /
• v.24 no.2
• /
• pp.221-241
• /
• 2018

Deep learning is getting attention recently. The deep learning technique which had been applied in competitions of the International Conference on Image Recognition Technology(ILSVR) and AlphaGo is Convolution Neural Network(CNN). CNN is characterized in that the input image is divided into small sections to recognize the partial features and combine them to recognize as a whole. Deep learning technologies are expected to bring a lot of changes in our lives, but until now, its applications have been limited to image recognition and natural language processing. The use of deep learning techniques for business problems is still an early research stage. If their performance is proved, they can be applied to traditional business problems such as future marketing response prediction, fraud transaction detection, bankruptcy prediction, and so on. So, it is a very meaningful experiment to diagnose the possibility of solving business problems using deep learning technologies based on the case of online shopping companies which have big data, are relatively easy to identify customer behavior and has high utilization values. Especially, in online shopping companies, the competition environment is rapidly changing and becoming more intense. Therefore, analysis of customer behavior for maximizing profit is becoming more and more important for online shopping companies. In this study, we propose 'CNN model of Heterogeneous Information Integration' using CNN as a way to improve the predictive power of customer behavior in online shopping enterprises. In order to propose a model that optimizes the performance, which is a model that learns from the convolution neural network of the multi-layer perceptron structure by combining structured and unstructured information, this model uses 'heterogeneous information integration', 'unstructured information vector conversion', 'multi-layer perceptron design', and evaluate the performance of each architecture, and confirm the proposed model based on the results. In addition, the target variables for predicting customer behavior are defined as six binary classification problems: re-purchaser, churn, frequent shopper, frequent refund shopper, high amount shopper, high discount shopper. In order to verify the usefulness of the proposed model, we conducted experiments using actual data of domestic specific online shopping company. This experiment uses actual transactions, customers, and VOC data of specific online shopping company in Korea. Data extraction criteria are defined for 47,947 customers who registered at least one VOC in January 2011 (1 month). The customer profiles of these customers, as well as a total of 19 months of trading data from September 2010 to March 2012, and VOCs posted for a month are used. The experiment of this study is divided into two stages. In the first step, we evaluate three architectures that affect the performance of the proposed model and select optimal parameters. We evaluate the performance with the proposed model. Experimental results show that the proposed model, which combines both structured and unstructured information, is superior compared to NBC(Naïve Bayes classification), SVM(Support vector machine), and ANN(Artificial neural network). Therefore, it is significant that the use of unstructured information contributes to predict customer behavior, and that CNN can be applied to solve business problems as well as image recognition and natural language processing problems. It can be confirmed through experiments that CNN is more effective in understanding and interpreting the meaning of context in text VOC data. And it is significant that the empirical research based on the actual data of the e-commerce company can extract very meaningful information from the VOC data written in the text format directly by the customer in the prediction of the customer behavior. Finally, through various experiments, it is possible to say that the proposed model provides useful information for the future research related to the parameter selection and its performance.

• Korean Journal of Environment and Ecology
• /
• v.34 no.1
• /
• pp.63-71
• /
• 2020

This study characterized the spatial and seasonal patterns of light pollution in the Deogyusan National Park and examined the potential effects of light pollution on ecosystems in the park using light intensities derived from VIIRS (Visible Infrared Imaging Radiometer Suite) DNB (Day and Night Band) nightlight images collected in January and August 2018. Results showed that the Muju Deogyusan resort had the greatest light intensity than other sources of light pollution in the park, and light intensity of the resort was much higher in January than in August, suggesting that artificial lights in ski slopes and facilities were the major source of light pollution in the park. An analysis of an urban-natural light pollution gradient along a neighboring urban area through the inside of the park indicated that light radiated from a light pollution source permeated for up to 1km into the adjacent area and contaminated the edge area of the park. Of the legally protected species whose distributions were reported in literature, four mammals (Martes flavigula, Mustela nivalis, Prionailurus bengalensis, Pteromys volans aluco), two birds (Falco subbuteo, Falco tinnunculus), and nine amphibians and reptiles (Onychodactylus koreanus, Hynobius leechii, Karsenia koreana, Rana dybowskii, Rana huanrenensis, Elaphe dione, Rhabdophis tigrinus, Gloydius ussuriensis, Gloydius saxatilis) inhabited light-polluted areas. Of those species inhabiting light-polluted areas, nocturnal species, such as Prionailurus bengalensis and Pteromys volans aluco, in particular, were vulnerable to light pollution. These results implied that protecting ecosystems from light pollution in national parks requires managing nighttime light in the parks and surrounding areas and making a plan to manage nighttime light pollution by taking into account ecological characteristics of wild animals in the parks.

• Journal of the Korean Academic Society of Industrial Cluster
• /
• v.1 no.1
• /
• pp.67-86
• /
• 2007

Wonju Medical Equipment Industry, despite of its short history, poor sales and weak manpower and so on, have shown remarkable outcomes in a relatively short period. At the end of 2007, totally 79 enterprises (only 4.6% of whole enterprises in Korea) made 10% of the nationwide production and 15% of the nationwide exports with an annual average growth rate of 66.7%, contributing domestic medical equipment industry tremendously. In addition, many leading medical equipment enterprises in various fields already moved or plan to move to Wonju, accelerating Wonju Medical Equipment Cluster. Wonju Medical Equipment Industry Cluster now enters into the growth stage, getting out of the initial business setup stage. Especially, the nomination of Wonju cluster project from the government accelerates networking (e.g. the development of the universal parts, the establishment of the mutual collaboration model among enterprises, and the mutual marketing), making a rapid growth in Wonju Medical Equipment Industry. Wonju Medical Equipment Industry Cluster revealed positive outcomes despite of the weakness in investment size and infra-structure comparing with the other medical industry cluster in the advanced country, while many domestic enterprises pursued their own growth models and thus failed to promote the international competitive power. Wonju Medical Equipment Industry has been developed rapidly. However, there are many challenging problems to support enterprises: small R&D investment and thus weak technology power, difficulties in recruiting R&D engineers, and poor marketing capabilities, financial infrastructure & policies, and network architecture. In order to develop a world-competitive medical equipment industry cluster at Wonju, the complement of infrastructures, the technology innovation, the mutual marketing, and the network expansion to support enterprises are further required. Wonju' s experiences in developing medical equipment industry so far suggest that our own flexible cluster model considering the industry structure and maturity for different regions should be developed, and specific action plans from the local and central governments based on their systematic strategies for industry development should be implemented in order to build world-competitive industry clusters in Korea.

• Journal of Korean Society of Forest Science
• /
• v.105 no.1
• /
• pp.149-156
• /
• 2016

Forest museums can be defined as facilities for the collection, exhibition, and education of the forest or forest related artifacts or data. This study was performed to measure the educational value of Gyeongnam state forest museum's forest and its environment. The tool used was the Contingent Valuation Methods (CVM) which is well known as a value estimation tool of environmental goods. The study for the value estimation is performed from April, 2014 to October of the same year through selection of the subject, decision of proposed price, and orientation of the survey staffs and total of 386 surveys were used in analysis. The value estimation tool used the DBDC logit model and the input parameters were number of visit (time), degree of environmental education (contri), the environment conservation effort of the respondent (execu), the education level of the respondent (edu), and income of the respondent (inc) and trimmed mean (WTPtruncated) was used. The estimated value of flora and environment education per each person per visit is 23,338 won. When applied to the average annual visitors deducted from 2010 to 2014, which is 430,000 per year, the environmental value that Gyeongnam state forest museum is providing to visitors each year is about 10 billion won. The result of this study is significant to propose the value of forest education and environment that the forest museum is offering to the visitors in the current currency. This is an evidence to directly determine the value of the forest museum and therefore proposing an opportunity change the recognition toward the forest and environment education.

• Economic and Environmental Geology
• /
• v.39 no.3 s.178
• /
• pp.235-253
• /
• 2006

We interpreted marine seismic profiles in conjunction with swath bathymetric and magnetic data to investigate rifting to breakup processes at the Korean margin leading to the separation of the Japan Arc. The Korean margin is rimmed by fundamental elements of rift architecture comprizing a seaward succession of a rift basin and an uplifted rift flank passing into the slope, typical of a passive continental margin. In the northern part, rifting occurred in the Korea Plateau, a continental fragment extended and partially segmented from the Korean Peninsula, that provided a relatively broader zone of extension resulting in a number of rifts. Two distinguished rift basins (Onnuri and Bandal Basins) in the Korea Plateau we bounded by major synthetic and smaller antithetic faults, creating wide and symmetric profiles. The large-offset border fault zones of these basins have convex dip slopes and demonstrate a zig-zag arrangement along strike. In contrast, the southern margin is engraved along its length with a single narrow rift basin (Hupo Basin) that is an elongated asymmetric half-graben. Rifting at the Korean margin was primarily controlled by normal faulting resulting from extension in the west and southeast directions orthogonal to the inferred line of breakup along the base of the slope rather than strike-slip deformation. Although rifting involved no significant volcanism, the inception of sea floor spreading documents a pronounced volcanic phase which seems to reflect slab-induced asthenospheric upwelling as well as rift-induced convection particularly in the narrow southern margin. We suggest that structural and igneous evolution of the Korean margin can be explained by the processes occurring at the passive continental margin with magmatism intensified by asthenospheric upwelling in a back-arc setting.

• Progress in Medical Physics
• /
• v.26 no.3
• /
• pp.143-152
• /
• 2015

In this study, we aim to design the architecture of the kV imaging system for tumor tracking in the dual-head gantry system and analyze its accuracy by simulations. We established mathematical formulas and algorithms to track the tumor position with the two-pair kV imaging systems when they are in the non-orthogonal positions. The algorithms have been designed in the homogeneous coordinate framework and the position of the source and the detector coordinates are used to estimate the tumor position. 4D XCAT (4D extended cardiac-torso) software was used in the simulation to identify the influence of the angle between the two-pair kV imaging systems and the resolution of the detectors to the accuracy in the position estimation. A metal marker fiducial has been inserted in a numerical human phantom of XCAT and the kV projections were acquired at various angles and resolutions using CT projection software of the XCAT. As a result, a positional accuracy of less than about 1mm was achieved when the resolution of the detector is higher than 1.5 mm/pixel and the angle between the kV imaging systems is approximately between $90^{\circ}$ and $50^{\circ}$. When the resolution is lower than 1.5 mm/pixel, the positional errors were higher than 1mm and the error fluctuation by the angles was greater. The resolution of the detector was critical in the positional accuracy for the tumor tracking and determines the range for the acceptable angle range between the kV imaging systems. Also, we found that the positional accuracy analysis method using XCAT developed in this study is highly useful and will be a invaluable tool for further refined design of the kV imaging systems for tumor tracking systems.

• Journal of the Korea Concrete Institute
• /
• v.22 no.6
• /
• pp.851-858
• /
• 2010

In this study, waste glass and bottom ash were used as basic materials in order to secure a recycling technology of by-products which was mostly discarded and reclaimed. In addition, because softening point of waste glass is less than $700^{\circ}C$ and bottom ash includes combustible material, it was possible to manufacture low-temperature sintering lightweight aggregates for energy saving at $800{\sim}900^{\circ}C$ that it is as much as 20~30% lower than sintering temperature of existing lightweight aggregates. Thermal conductivity of newly-developed lightweight aggregates was 0.056~0.105W/m. K and its porosity was 40.36~84.89%. A coefficient of correlation between thermal conductivity and porosity was -0.97, it showed very high negative correlationship. With this, we were able to verify that porosity is key factor to affect thermal conductivity. Microstructure of lightweight aggregates by $CaCO_3$ content and replacement ratio of bottom ash in the variation of temperature were that $CaCO_3$ content increased along with pore size while replacement ratio of bottom ash increased as pore size decreased. Specially, most pores were open pore instead of closed pore of globular shape when replacement ratio of bottom ash was 30%, and pore size was small about 1/10~1/5 as compared with case in bottom ash 0~20%. In addition, open pore shapes were remarkably more irregular form of open pore in $900^{\circ}C$ than $700^{\circ}C$ or $800^{\circ}C$ when replacement ratio of bottom ash was 30%. We reasoned hereby that these results will influence on absorption increase, strength and thermal conductivity decrease of lightweight aggregates.

• Korean Journal of Environment and Ecology
• /
• v.28 no.2
• /
• pp.235-246
• /
• 2014

The purpose of this study was to determine the correlation between botanical garden visitors' ecological consciousness and their needs, in order to provide some effective measures to manage them. For this purpose, 3 study points were set up: "botanical garden visitors' ecological consciousness and their needs", "differences of such consciousness depending on their demographic variables" and the "relationship between such consciousness and their needs". To this end, Botanical garden visitors were surveyed for an empirical analysis. The visitors' awareness about ecology was measured with Dunlap's 15-item NEP Inventory, while their needs were analyzed in reference to Maslow's 7-Step Human Desire Ladder. The survey was conducted at Botanical garden for 3 days. As a result, a total of 360 questionnaires were returned. The results of this study can be summarized as follows; First, the visitors' ecological consciousness and their needs were higher than normal level. In terms of their consciousness of ecology, their awareness of the ecological crisis potential and anti-humanism were the highest. In terms of their needs, the aesthetic need was the highest, followed by the cognitive need. On the other hand, the needs for self-achievement and self-esteem were the lowest; except them, the higher the needs were positioned at Maslow's ladder of desire, the more responsive the subjects became. As a result of analyzing the correlation between the subjects' consciousness of ecology and their needs, it was found that the correlation was negative in some sub-areas, while being positive in other sub-areas. After all, the ratio of the sub-areas having a positive correlation was 3 times higher than that of the sub-areas having a negative correlation. Even as for the correlation coefficient values, they were higher in the positive sub-areas, which suggests that the correlation between wetland visitors' ecological consciousness and their needs was positive, although at a lower level, in overall terms. As a result of comparatively analyzing visitors' needs by dividing them into 3 sub-groups depending on the levels of their ecological consciousness, it was found that the higher their consciousness of ecology was, the higher their needs were. Overall, botanical garden visitors' ecological awareness was higher than the normal level, and it was estimated that such awareness would continue to increase. Hence, it could be inferred that their needs, particularly their aesthetic and cognitive ones, would also continue to increase. Accordingly, it is important to manage the wetland landscape making use of its visual resources, while keep providing the visitors with the contents fulfilling their need for knowledge.

• Journal of Environmental Policy
• /
• v.11 no.3
• /
• pp.25-47
• /
• 2012

This research is to suggest an ecological management methods by classification of biotope type and its mapping for Tricholoma matsutake. The target area was construction site of the East Hongchun - Yangyang Highway (Yangyang section) where massive production of Tricholoma matsutakes is in place. Specifically, the 700 m long range of area around the road route, Yangyang section, which is over a total area of $19,79km^2$. The flow of this research was as follows: first studied were the characteristics of the Yangyang Tricholoma matsutakes; second, reviews on previous studies on Tricholoma matsutakes; third, the formulation of standards and processes of biotope type considering Tricholoma matsutake production; and fourth, mapping of biotope considering the production area of Tricholoma matsutake. In 24.77% of the region, slope angle measured $30{\sim}40^{\circ}C$, which is suitable for Tricholoma matsutake production. Also, 17.44% of the region had southward vegetation structures, and as for the Pinus densiflora; 26.00% had average receipts; the tree ages was $38{\pm}8.34$ years; and the average density of canopy layer was $9.55{\pm}4.89/100m^2$. The soil acidity (pH) values of 5.0 to 5.6(47.96%) > 5.6 to 7.0(42.90%) > 4.0 to 5.0(9.14%) were slightly high for Tricholoma matsutake growth; but 57.10% of the area had a soil pH of 4.0-5.6. The organic horizon (A0 layer) was $3.39{\pm}2.14cm$, and the mean depths were 4-6cm(78.03%) > 0-2cm(18.10%) > 2-4cm(3.87%). Based on the results of previous studies and field researches on the presence of vegetation (the pine dominance ratio), the topographic structure (the terrain characteristics and slopes), the receipt, the soil structure (the soil acidity and the organic layers interposed between the depths), and the vegetation density, six clusters based on the identified types of biotopes are produced in this paper. Area of distribution according to the type of biotope was Tricholoma matsutake of the potential-production (II) biotope (32.86%) > Tricholoma matsutake of the inadequate-production biotope (22.17%) > Tricholoma matsutake of the adequate (II)-production biotope (17.79%) > Tricholoma matsutake of the adequate (I)-production biotope (14.86%) > Tricholoma matsutake of the potential-production (I) biotope (9.77%) > Tricholoma matsutake of the produced biotope (2.55%).

• Sleep Medicine and Psychophysiology
• /
• v.12 no.2
• /
• pp.117-121
• /
• 2005

Objectives: The purpose of this study was to assess the difference of subjective daytime sleepiness level between primary insomnia patients and healthy control subjects. We also investigated the relationship between subjective daytime sleepiness level and variables of nocturnal polysomnograghic sleep architecture of insomnia patients. Method: Total subjects were 87 patients with primary insomnia diagnosed with polysomnography and 88 normal controls. The daytime sleepiness level in each group was measured by Korean version of Epworth Sleepiness Scale (ESS). The correlations of ESS score and nocturnal polysomnographic variables were calculated in the patient group. Results: Patients with insomnia had the lower ESS scores than the control group. In patients group, the ESS score showed significant negative correlations with total sleep time, sleep efficiency%, and stage 2 sleep time%. The ESS score also showed significant positive correlations with number of awakenings, number of awakenings more than 2 minutes, and wake after sleep onset time. Conclusions: Insomnia patients showed lower level of subjective daytime sleepiness that may indicate their higher alertness comparing to control subjects. Daytime sleepiness of patients with insomnia was associated with polysomnographic variables including total sleep time, sleep efficiency%, stage 2 sleep time% and disrupted continuity of nocturnal sleep.