• Journal of the Korean association of regional geographers
• /
• v.3 no.2
• /
• pp.37-87
• /
• 1997

As a results of researches on the cultivation processes and settlement developments on the Mangyoung river valley as a whole could be have four 'Space-Time Continuity' through a [Origin-Destination] theory model. On a initial phases of cultivation, the cultivation process has been begun at mountain slopes and tributory plains in upper part of river-basin from Koryo Dynasty to early Chosun Dynasty. At first, indigenous peasants burned forests on the mountain slopes for making 'dryfield' for a cereal crops. Following population increase more stable food supply is necessary facets of life inducing a change production method into a 'wetfield' in tributory plains matching the population increase. First sedentary agriculture maybe initiated at this mountain slopes and tributory plains on upper part of river basin through a burning cultivation methods. Mountain slopes and tributory plains are become a Origin area in cultivation processes. It expanded from up to down through the valleys with 'a bits of land' fashion in a steady pace like a terraced fields expanded with bit by bit of land to downward. They expanded their land to the middle part of river basin in mid period of Chosun Dynasty with dike construction techniques on the river bank. Lower part of river cultivated with embankment building techniques in 1920s and then naturally expanded to the tidal marshes on the estuaries and river inlets of coastal areas. 'Pioneer fringes' are consolidated at there in modern times. Changes in landscapes are appeared it's own characters with each periods of time. Followings are results of study through the Mangyoung river valley as a whole. (1) Mountain slopes and tributory plains on the upper part of river are cultivated 'dryfields' by indigenous peasants with Burning cultivation methods at first and developed sedentary settlements at the edges of mountain slopes and on the river terrace near the fields. They formed a kind of 'periphery-located cluster type' of settlement. This type of settlement are become a prominant type in upper part of river basin. 'Dryfields' has been changed into a 'wetfields' at the narrow tributory plains by increasing population pressure in later time. These wetfields are supplied water by Weir and Ponds Irrigation System(제언수리방법). Streams on the tributory plains has been attracted wetfields besides of it and formed a [water+land] complex on it. 'Wetfields' are expanded from up to downward with a terraced land pattern(adder like pattern, 붕전) according to the gradient of valley. These periphery located settlements are formed a intimate ecological linkage with several sets of surroundings. Inner villages are expanded to Outer villages according to the expansion of arable lands into downward. (2) Mountain slopes and tributory plains expanded its territory to the alluvial deposited plains on the middle part of river valley with a urgent need of new land by population increase. This part of alluvial plains are cultivated mainly in mid period of Chosun Dynasty. Irrigation methods are changed into a Dike Construction Irrigation method(천방수리방법) for the control of floods. It has a trend to change the subjectives of cultivation from community-oriented one who constructed Bochang along tributories making rice paddies to local government authorities who could be gather large sums of capitals, techniques and labours for the big dike construction affairs. Settlements are advanced in the midst of plains avoiding friction of distances and formed a 'Centrallocated cluster type' of settlements. There occured a hierarchical structures of settlements in ranks and sizes according merits of water supply and transportation convenience at the broad plains. Big towns are developed at there. It strengthened a more prominant [water+land] complex along the canals. Ecological linkages between settlements and surroundings are shaded out into a tiny one in this area. (3) It is very necessary to get a modern technology of flood control at the rivers that have a large volume of water and broad width. The alluvial plains are remained in a wilderness phase until a technical level reached a large artificial levee construction ability that could protect the arable land from flood. Until that time on most of alluvial land at the lower part of river are remained a wilderness of overgrown with reeds in lacks of techniques to build a large-scale artificial levee along the riverbank. Cultivation processes are progressed in a large scale one by Japanese agricultural companies with [River Rennovation Project] of central government in 1920s. Large scale artificial levees are constructed along the riverbank. Subjectives of cultivation are changed from Korean peasants to Japanese agricultural companies and Korean peasants fell down as a tenant in a colonial situation of that time in Korea. They could not have any voices in planning of spatial structure and decreased their role in planning. Newly cultivated lands are reflected company's intensions, objectives and perspectives for achieving their goals for the sake of colonial power. Newly cultivated lands are planned into a regular Rectangular Block settings of rice paddies and implanted a large scale Bureaucratic-oriented Irrigation System on the cultivated plains. Every settlements are located in the midst of rice paddies with a Central located Cluster type of settlements. [water+land] complex along the canal system are more strengthened. Cultivated space has a characters of [I-IT] landscapes. (4) Artificial levees are connected into a coastal emnankment for a reclamation of broad tidal marshes on the estuaries and inlets of rivers in the colonial times. Subjectives of reclamation are enlarged into a big agricultural companies that could be acted a role as a big cultivator. After that time on most of reclamation project of tidal marshes are controlled by these agricultural companies formed by mostly Japanese capitalists. Reclaimed lands on the estuaries and river inlets are under hands of agricultural companies and all the spatial structures are formed by their intensions, objectives and perspectives. They constructed a Unit Farming Area for the sake of companies. Spatial structures are planned in a regular one with broad arable land for the rice production of rectangular blocks, regular canal systems and tank reservoir for the irrigation water supply into reclaimed lands. There developed a 'Central-located linear type' of settlements in midst of reclaimed land. These settlements are settled in a detail program upon this newly reclaimed land at once with a master plan and they have planned patterns in their distribution, building materials, location, and form. Ecological linkage between Newly settled settlemrnts and its surroundings are lost its colours and became a more artificial one by human-centred environment. [I-IT] landscapes are become more prominant. This region is a destination area of [Origin-Destination] theory model and formed a 'Pioneer Fringe'. It is a kind of pioneer front that could advance or retreat discontinously by physical conditions and socio-cultural conditions of that region.

• Korean Journal of Ecology and Environment
• /
• v.47 no.2
• /
• pp.91-102
• /
• 2014

Forest vegetation of Birobong (1,563 m) in Odaesan National Park is classified into mountain forest vegetation and flatland forest vegetation. Mountain forest vegetation is subdivided into deciduous broad-leaved forest, valley forest, coniferous forest, subalpine coniferous forest, subalpine broad-leaved forest, afforestation and etc., while riparian forest was found under the category of flatland forest vegetation. Including 196 communities of mountain forest vegetation, 1 community of flatland forest vegetation and 4 communities of other vegetation, the total of 201 communities were researched; the distributed colonies classified by physiognomy classification are 62 communities deciduous broad-leaved forest, 84 communities of valley forest, 15 communities of coniferous forests, 16 communities of subalpine coniferous forest, 3 communities of subalpine broad-leaved forest, 16 afforestation, 1 community of flatland forest and 4 other communities. As for the distribution rate for surveyed main communities, Quercus mongolica, Quercus variabilis, Tilia amurensis communities account for 37.08 percent of deciduous broad-leaved forest, Juglans mandshurica, Fraxinus mandshurica, Cornus controversa, Populus koreana community takes up 1.59 percent of mountain valley forest, Pinus densiflora community holds 6.65 percent of mountain coniferous forest holds. In conclusion, minority species consisting of Quercus mongolica, Quercus variabilis, Tilia amurensis, Juglans mandshurica, Fraxinus mandshurica, Cornus controversa, Populus koreana, Pinus densiflora are distributed as dominant species of the uppermost part in a forest vegetation of Birobong in Odaesan National Park. In addition, because of vegetation succession and climate factors, numerous colonies formed by the two species are expected to be replaced by Quercus mongolica, Tilia amurensis and Juglans mandshurica which are climax species in the area. However, the distribution rate of deciduous broad-leaved forest seems to increase gradually due to global warming and artificial disturbance.

• Korean Journal of Environment and Ecology
• /
• v.26 no.2
• /
• pp.219-232
• /
• 2012

The purpose of this study was to find characteristics of $Pinus$ $densiflora$ forests in previous records and cultural landscape regions and to establish a management method for recovery of $P.$ $densiflora$ landscape. A total of 16 landscape paintings were analyzed to study $P.$ $densiflora$ forests in Inwangsan(Mt.) shown in figure data. As a result of analyzing figure data, $P.$ $densiflora$ $forests$ were found to be distributed mainly in the ridge line of east slope side of Inwangsan(Mt.) from landscape paintings of Jung Seon, Jang Si Heung and Gang Hee Un. In order to analyze the landscape of $P.$ $densiflora$ forests shown in photograph data on Inwangsan(Mt.), photographs of Inwangsan(Mt.) since late 1800s were searched and a total of 24 photographs were used for analysis. As a result of photographic analysis, $P.$ $densiflora$ forests were found to be distributed mainly in the low area and ridge line of western slope side of Inwangsan(Mt.). As a result of analyzing the distribution status of 10 cultural heritages of Inwangsan(Mt.), cultural heritages in the mountain were mainly established with rocks and $P.$ $densiflora$ forests as the background and are establishing their landscape not only as a form of cultural heritage but also including the natural landscape in the background. A method of restoring cultural landscape of P. densiflora forests was necessary. As a result of analyzing 26 rock beds of Inwangsan(Mt.), most of $P.$ $densiflora$ forests were distributed nearby rock beds since the past. Among such regions, restoration and management of introduced species in the accumulated soil regions to $P.$ $densiflora$ forests were deemed necessary. As a result of analyzing current vegetation of $P.$ $densiflora$ forest areas of the past, 51.87% of areas maintained $P.$ $densiflora$ forest and 25.80% were changed to artificial forests. Most of low areas of the mountain were turned into urban area (10.16%). Management type for restoration of $P.$ $densiflora$ landscape of Inwangsan(Mt.) was classified into preservation, $P.$ $densiflora$ restoration and maintenance. Preservation management was configured as $P.$ $densiflora$ distribution and rocky area distribution. The targets of $P.$ $densiflora$ restoration were indigenous broad-leaf trees, vegetation areas with introduced species and damaged regions within the mountain. The targets of maintenance were urban area in the low area of Inwangsan(Mt.) and military base at the top of the mountain.

• Korean Journal of Ecology and Environment
• /
• v.46 no.3
• /
• pp.449-459
• /
• 2013

Forest vegetation of Geochilbong (1,177 m) in Deogyusan National Park is classified into mountain forest vegetation and flatland forest vegetation. Mountain forest vegetation is subdivided into deciduous broad-leaved forest, valley forest, coniferous forest, afforestation and etc., while riparian forest was found under the category of flatland forest vegetation. Including 89 communities of mountain forest vegetation and 4 communities of other vegetation, the total of 93 communities were researched; the distributed colonies classified by physiognomy classification are 32 communities deciduous broadleaved forest, 21 communities of valley forest, 12 communities of coniferous forests, 24 afforestation and 4 other communities. As for the distribution rate for surveyed main communities, Quercus mongolica, Quercus serrata, Quercus variabilis communities account for 56.54 percent of deciduous broad-leaved forest, Fraxinus mandshurica, Cornus controversa community takes up 46.58 percent of mountain valley forest, Pinus densiflora community holds 74.98 percent of mountain coniferous forest holds. In conclusion, minority species consisting of Quercus mongolica, Quercus serrata, Quercus variabilis, Fraxinus mandshurica, Cornus controversa, Pinus densiflora are distributed as dominant species of the uppermost part in a forest vegetation of Geochilbong in Deogyusan National Park. In addition, because of vegetation succession and climate factors, numerous colonies formed by the two species are expected to be replaced by Quercus mongolica, Carpinus laxiflora and Fraxinus mandshurica which are climax species in the area. However, the distribution rate of deciduous broad-leaved forest seems to increase gradually due to global warming and artificial disturbance.

• Journal of Intelligence and Information Systems
• /
• v.27 no.3
• /
• pp.57-73
• /
• 2021

Maintenance and prevention of failure through anomaly detection of ICT infrastructure is becoming important. System monitoring data is multidimensional time series data. When we deal with multidimensional time series data, we have difficulty in considering both characteristics of multidimensional data and characteristics of time series data. When dealing with multidimensional data, correlation between variables should be considered. Existing methods such as probability and linear base, distance base, etc. are degraded due to limitations called the curse of dimensions. In addition, time series data is preprocessed by applying sliding window technique and time series decomposition for self-correlation analysis. These techniques are the cause of increasing the dimension of data, so it is necessary to supplement them. The anomaly detection field is an old research field, and statistical methods and regression analysis were used in the early days. Currently, there are active studies to apply machine learning and artificial neural network technology to this field. Statistically based methods are difficult to apply when data is non-homogeneous, and do not detect local outliers well. The regression analysis method compares the predictive value and the actual value after learning the regression formula based on the parametric statistics and it detects abnormality. Anomaly detection using regression analysis has the disadvantage that the performance is lowered when the model is not solid and the noise or outliers of the data are included. There is a restriction that learning data with noise or outliers should be used. The autoencoder using artificial neural networks is learned to output as similar as possible to input data. It has many advantages compared to existing probability and linear model, cluster analysis, and map learning. It can be applied to data that does not satisfy probability distribution or linear assumption. In addition, it is possible to learn non-mapping without label data for teaching. However, there is a limitation of local outlier identification of multidimensional data in anomaly detection, and there is a problem that the dimension of data is greatly increased due to the characteristics of time series data. In this study, we propose a CMAE (Conditional Multimodal Autoencoder) that enhances the performance of anomaly detection by considering local outliers and time series characteristics. First, we applied Multimodal Autoencoder (MAE) to improve the limitations of local outlier identification of multidimensional data. Multimodals are commonly used to learn different types of inputs, such as voice and image. The different modal shares the bottleneck effect of Autoencoder and it learns correlation. In addition, CAE (Conditional Autoencoder) was used to learn the characteristics of time series data effectively without increasing the dimension of data. In general, conditional input mainly uses category variables, but in this study, time was used as a condition to learn periodicity. The CMAE model proposed in this paper was verified by comparing with the Unimodal Autoencoder (UAE) and Multi-modal Autoencoder (MAE). The restoration performance of Autoencoder for 41 variables was confirmed in the proposed model and the comparison model. The restoration performance is different by variables, and the restoration is normally well operated because the loss value is small for Memory, Disk, and Network modals in all three Autoencoder models. The process modal did not show a significant difference in all three models, and the CPU modal showed excellent performance in CMAE. ROC curve was prepared for the evaluation of anomaly detection performance in the proposed model and the comparison model, and AUC, accuracy, precision, recall, and F1-score were compared. In all indicators, the performance was shown in the order of CMAE, MAE, and AE. Especially, the reproduction rate was 0.9828 for CMAE, which can be confirmed to detect almost most of the abnormalities. The accuracy of the model was also improved and 87.12%, and the F1-score was 0.8883, which is considered to be suitable for anomaly detection. In practical aspect, the proposed model has an additional advantage in addition to performance improvement. The use of techniques such as time series decomposition and sliding windows has the disadvantage of managing unnecessary procedures; and their dimensional increase can cause a decrease in the computational speed in inference.The proposed model has characteristics that are easy to apply to practical tasks such as inference speed and model management.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.83-102
• /
• 2021

The government recently announced various policies for developing big-data and artificial intelligence fields to provide a great opportunity to the public with respect to disclosure of high-quality data within public institutions. KSURE(Korea Trade Insurance Corporation) is a major public institution for financial policy in Korea, and thus the company is strongly committed to backing export companies with various systems. Nevertheless, there are still fewer cases of realized business model based on big-data analyses. In this situation, this paper aims to develop a new business model which can be applied to an ex-ante prediction for the likelihood of the insurance accident of credit guarantee. We utilize internal data from KSURE which supports export companies in Korea and apply machine learning models. Then, we conduct performance comparison among the predictive models including Logistic Regression, Random Forest, XGBoost, LightGBM, and DNN(Deep Neural Network). For decades, many researchers have tried to find better models which can help to predict bankruptcy since the ex-ante prediction is crucial for corporate managers, investors, creditors, and other stakeholders. The development of the prediction for financial distress or bankruptcy was originated from Smith(1930), Fitzpatrick(1932), or Merwin(1942). One of the most famous models is the Altman's Z-score model(Altman, 1968) which was based on the multiple discriminant analysis. This model is widely used in both research and practice by this time. The author suggests the score model that utilizes five key financial ratios to predict the probability of bankruptcy in the next two years. Ohlson(1980) introduces logit model to complement some limitations of previous models. Furthermore, Elmer and Borowski(1988) develop and examine a rule-based, automated system which conducts the financial analysis of savings and loans. Since the 1980s, researchers in Korea have started to examine analyses on the prediction of financial distress or bankruptcy. Kim(1987) analyzes financial ratios and develops the prediction model. Also, Han et al.(1995, 1996, 1997, 2003, 2005, 2006) construct the prediction model using various techniques including artificial neural network. Yang(1996) introduces multiple discriminant analysis and logit model. Besides, Kim and Kim(2001) utilize artificial neural network techniques for ex-ante prediction of insolvent enterprises. After that, many scholars have been trying to predict financial distress or bankruptcy more precisely based on diverse models such as Random Forest or SVM. One major distinction of our research from the previous research is that we focus on examining the predicted probability of default for each sample case, not only on investigating the classification accuracy of each model for the entire sample. Most predictive models in this paper show that the level of the accuracy of classification is about 70% based on the entire sample. To be specific, LightGBM model shows the highest accuracy of 71.1% and Logit model indicates the lowest accuracy of 69%. However, we confirm that there are open to multiple interpretations. In the context of the business, we have to put more emphasis on efforts to minimize type 2 error which causes more harmful operating losses for the guaranty company. Thus, we also compare the classification accuracy by splitting predicted probability of the default into ten equal intervals. When we examine the classification accuracy for each interval, Logit model has the highest accuracy of 100% for 0~10% of the predicted probability of the default, however, Logit model has a relatively lower accuracy of 61.5% for 90~100% of the predicted probability of the default. On the other hand, Random Forest, XGBoost, LightGBM, and DNN indicate more desirable results since they indicate a higher level of accuracy for both 0~10% and 90~100% of the predicted probability of the default but have a lower level of accuracy around 50% of the predicted probability of the default. When it comes to the distribution of samples for each predicted probability of the default, both LightGBM and XGBoost models have a relatively large number of samples for both 0~10% and 90~100% of the predicted probability of the default. Although Random Forest model has an advantage with regard to the perspective of classification accuracy with small number of cases, LightGBM or XGBoost could become a more desirable model since they classify large number of cases into the two extreme intervals of the predicted probability of the default, even allowing for their relatively low classification accuracy. Considering the importance of type 2 error and total prediction accuracy, XGBoost and DNN show superior performance. Next, Random Forest and LightGBM show good results, but logistic regression shows the worst performance. However, each predictive model has a comparative advantage in terms of various evaluation standards. For instance, Random Forest model shows almost 100% accuracy for samples which are expected to have a high level of the probability of default. Collectively, we can construct more comprehensive ensemble models which contain multiple classification machine learning models and conduct majority voting for maximizing its overall performance.

• Journal of Intelligence and Information Systems
• /
• v.17 no.3
• /
• pp.25-41
• /
• 2011

Recently, support vector machines (SVMs) are being recognized as competitive tools as compared with other data mining techniques for solving pattern recognition or classification decision problems. Furthermore, many researches, in particular, have proved them more powerful than traditional artificial neural networks (ANNs) (Amendolia et al., 2003; Huang et al., 2004, Huang et al., 2005; Tay and Cao, 2001; Min and Lee, 2005; Shin et al., 2005; Kim, 2003).The classification decision, such as a binary or multi-class decision problem, used by any classifier, i.e. data mining techniques is so cost-sensitive particularly in financial classification problems such as the credit ratings that if the credit ratings are misclassified, a terrible economic loss for investors or financial decision makers may happen. Therefore, it is necessary to convert the outputs of the classifier into wellcalibrated posterior probabilities-based multiclass credit ratings according to the bankruptcy probabilities. However, SVMs basically do not provide such probabilities. So it required to use any method to create the probabilities (Platt, 1999; Drish, 2001). This paper applied AdaBoost algorithm-based support vector machines (SVMs) into a bankruptcy prediction as a binary classification problem for the IT companies in Korea and then performed the multi-class credit ratings of the companies by making a normal distribution shape of posterior bankruptcy probabilities from the loss functions extracted from the SVMs. Our proposed approach also showed that their methods can minimize the misclassification problems by adjusting the credit grade interval ranges on condition that each credit grade for credit loan borrowers has its own credit risk, i.e. bankruptcy probability.

• Journal of Korea Water Resources Association
• /
• v.47 no.6
• /
• pp.537-546
• /
• 2014

Surface image velocimetry was introduced as an efficient and sage alternative to conventional river flow measurement methods during floods. The conventional surface image velocimetry uses a pair of images to estimate velocity fields using cross-correlation analysis. This method is appropriate to analyzing images taken with a short time interval. It, however, has some drawbacks; it takes a while to analyze images for the verage velocity of long time intervals and is prone to include errors or uncertainties due to flow characteristics and/or image taking conditions. Methods using spatio-temporal images, called STIV, were developed to overcome the drawbacks of conventional surface image velocimetry. The grayscale-gradient tensor method, one of various STIVs, has shown to be effectively reducing the analysis time and is fairly insusceptible to any measurement noise. It, unfortunately, can only be applied to the main flow direction. This means that it can not measure any two-dimensional flow field, e.g. flow in the vicinity of river structures and flow around river bends. The present study aimed to develop a new method of analyzing spatio-temporal images in two-dimension using cross-correlation analysis. Unlike the conventional STIV, the developed method can be used to measure two-dimensional flow substantially. The method also has very high spatial resolution and reduces the analysis time. A verification test using artificial images with lid-driven cavity flow showed that the maximum error of the method is less than 10 % and the average error is less than 5 %. This means that the developed scheme seems to be fairly accurate, even for two-dimensional flow.

• Journal of Embryo Transfer
• /
• v.31 no.1
• /
• pp.47-52
• /
• 2016

The objective of this study was to investigate the result of in vivo embryo collection and pregnancy rate after embryo transfer using sex-sorted sperm of Korean brindle cattle. Donor Korean brindle cattle superovulation treated by decreasing dose of FSH injection. Embryos were recovered on 7 days after the third artificial insemination. Control group semen straw used artificial insemination contained 20 million sperm. Sex-sorted semen straws contained 4 million sperm or 10 million sperm. As for the result of the recovery of the in vivo embryos derived from sex-sorted sperm, the number of transferable embryos was significantly highly recovered to be $6.20{\pm}2.28/donor$ from the control group and was significantly lowly recovered to be $1.57{\pm}1.72/donor$ from the group treated at a sperm concentration of $10{\times}10^6$ (p<0.05). The number of unfertilized embryo was $0.8{\pm}1.30/donor$ in control group which was significantly lower than the group treated at a sperm concentration of $4{\times}10^6$ (p<0.05). However, there was no significant difference in the number of undeveloped ova between control and treatment groups. Pregnancy rate after embryo transfer was shown to be 35.00% in control group and 12.50% in treatment group. The karyotype analysis of the calf derived from sex-sorted sperm resulted in a similar chromosomal distribution pattern (2n=60, XX) compared to those of common Korean native cattle.

• Journal of Korean Society of Forest Science
• /
• v.107 no.4
• /
• pp.351-360
• /
• 2018

This study was conducted to analyze the responses of chlorophyll contents and growth of Pinus densiflora and Quercus variabilis seedlings on distance from the well and $CO_2$ flux after the artificial $CO_2$ release. From June 1 to 30, 2016, $CO_2$ gas was injected at the rate of $6L\;min^{-1}$ at the study site in Eumseong. Chlorophyll content was analyzed in the middle of July, 2016, and root collar diameter (RCD), height (H), and biomass were measured in May and December, 2016 after planting 2-year-old P. densiflora and 1-year-old Q. variabilis seedlings in May, 2015. The chlorophyll content of P. densiflora seedlings did not show a significant correlation with $CO_2$ flux, whereas the chlorophyll content of Q. variabilis seedlings showed a significant negative correlation with increasing $CO_2$ flux (P<0.05). The RCD and H growth rates of both species showed the significant difference in the distance from the well of the $CO_2$ anthropogenic release treatment. In particular, the RCD and H growth rate of P. densiflora seedlings and the RCD growth rate of Q. variabilis seedlings increased significantly as the seedlings were closer to the well, but the H growth rate of Q. variabilis seedlings decreased significantly. In addition, as the $CO_2$ concentration in the ground increases, ${\Delta}R/S$ ratio increases in both species, suggesting that the high $CO_2$ concentration in the soil promotes carbon distribution relative to the root part. The results of this study can be used as data necessary to monitor the $CO_2$ leakage and physiological and growth responses of both species to leakage of stored $CO_2$ in the future.