• Journal of Intelligence and Information Systems
• /
• v.17 no.4
• /
• pp.157-173
• /
• 2011

As the Internet use explodes recently, the malicious attacks and hacking for a system connected to network occur frequently. This means the fatal damage can be caused by these intrusions in the government agency, public office, and company operating various systems. For such reasons, there are growing interests and demand about the intrusion detection systems (IDS)-the security systems for detecting, identifying and responding to unauthorized or abnormal activities appropriately. The intrusion detection models that have been applied in conventional IDS are generally designed by modeling the experts' implicit knowledge on the network intrusions or the hackers' abnormal behaviors. These kinds of intrusion detection models perform well under the normal situations. However, they show poor performance when they meet a new or unknown pattern of the network attacks. For this reason, several recent studies try to adopt various artificial intelligence techniques, which can proactively respond to the unknown threats. Especially, artificial neural networks (ANNs) have popularly been applied in the prior studies because of its superior prediction accuracy. However, ANNs have some intrinsic limitations such as the risk of overfitting, the requirement of the large sample size, and the lack of understanding the prediction process (i.e. black box theory). As a result, the most recent studies on IDS have started to adopt support vector machine (SVM), the classification technique that is more stable and powerful compared to ANNs. SVM is known as a relatively high predictive power and generalization capability. Under this background, this study proposes a novel intelligent intrusion detection model that uses SVM as the classification model in order to improve the predictive ability of IDS. Also, our model is designed to consider the asymmetric error cost by optimizing the classification threshold. Generally, there are two common forms of errors in intrusion detection. The first error type is the False-Positive Error (FPE). In the case of FPE, the wrong judgment on it may result in the unnecessary fixation. The second error type is the False-Negative Error (FNE) that mainly misjudges the malware of the program as normal. Compared to FPE, FNE is more fatal. Thus, when considering total cost of misclassification in IDS, it is more reasonable to assign heavier weights on FNE rather than FPE. Therefore, we designed our proposed intrusion detection model to optimize the classification threshold in order to minimize the total misclassification cost. In this case, conventional SVM cannot be applied because it is designed to generate discrete output (i.e. a class). To resolve this problem, we used the revised SVM technique proposed by Platt(2000), which is able to generate the probability estimate. To validate the practical applicability of our model, we applied it to the real-world dataset for network intrusion detection. The experimental dataset was collected from the IDS sensor of an official institution in Korea from January to June 2010. We collected 15,000 log data in total, and selected 1,000 samples from them by using random sampling method. In addition, the SVM model was compared with the logistic regression (LOGIT), decision trees (DT), and ANN to confirm the superiority of the proposed model. LOGIT and DT was experimented using PASW Statistics v18.0, and ANN was experimented using Neuroshell 4.0. For SVM, LIBSVM v2.90-a freeware for training SVM classifier-was used. Empirical results showed that our proposed model based on SVM outperformed all the other comparative models in detecting network intrusions from the accuracy perspective. They also showed that our model reduced the total misclassification cost compared to the ANN-based intrusion detection model. As a result, it is expected that the intrusion detection model proposed in this paper would not only enhance the performance of IDS, but also lead to better management of FNE.

• Korean Journal of Environment and Ecology
• /
• v.30 no.3
• /
• pp.386-398
• /
• 2016

To understand the vegetation structure in the ridgeline area of Nakdong-jeongmaek, six primary areas in Nakdong-jeongmaek were selected and their vegetation distribution was surveyed considering the environmental conditions and artificial influences. According to the results of community classification based on TWINSPAN, the vegetation in the surveyed region was categorized into 9 groups: Betula costata-Quercus mongolica community, Q. mongolica community, Pinus densiflora-Q. mongolica community, P. densiflora community, Deciduous oaks-P. densiflora community, Deciduous oaks community, P. thunbergii community, P. koraiensis-P. rigida community, and Chamaecyparis obtusa-Alnus firma community. In Baekbyeongsan(Mt.) located in Taebaek-si of Gangwon-do, Betula costata-Quercus mongolica community was found, reflecting the environmental characteristics of northern temperate climate. P. thunbergii community appeared in Gudeoksan(Mt.) of Busan metropolitan city, which is near the coast. Since Gudeoksan(Mt.) is near to the downtown and its altitude above the sea is relatively low, people visit the area often. Therefore, C. obtusa and P. thunbergii have been planted for producing forest trees and implementing anti-erosion afforestation. In the other primary survey areas, Q. mongolica-dominant communities, P. densiflora-dominant communities, and deciduous oak-dominant communities, which are representative forest vegetation types of Jeongmaeks in South Korea, were mainly distributed, showing no significant difference compared to the forest vegetation types of other Jeongmaeks. Since the Nakdong-jeongmaek from south to north, it shows clear characteristics of vegetation changes between the northern temperate climate and the warm temperate climate of the south.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.44 no.1
• /
• pp.40-50
• /
• 2016

A BIM, which is being applied actively to the construction and civil construction industries, is a technology that can maximize efficiency of various sectors from initial planning and design, construction, and maintenance, to demolition; however, it is in the introductory phase in the field of domestic landscaping. In order to introduce and promote BIM in the field of landscape design, this study developed a prototype of a library and template and analyzed the performance of trial application. For the development of a prototype, annotations and types were analyzed from floor plans of existing small parks, and components of landscape template were deduced. Based on this, play facilities, pergola, and benches were madeintofamily and templates, making automatic design possible. In addition, annotations and tags that are often used in landscape design were made, and a 3D view was materialized through visibility/graphic reassignment. As for tables and quantities, boundary stone table, mounding table, summary sheet of quantities, table of contents, and summary sheet of packaging quantities were grouped and connected with floor plans; regarding landscaping trees, classification criteria and name of trees that are suitable for domestic situations were applied. A landscape template was created to enable the library file format(rfa) that can be mounted on a building with BIM programs. As for problems that arose after the trial application of the prepared template, some CAD files could not be imported; also, while writing tables, the basis of calculation could not be made automatically. Regarding this, it is thought that functions of a BIM program and template need improvement.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.27 no.3
• /
• pp.1-17
• /
• 1999

The purpose of this study is to evaluate the creation techniques of artificial wetland, one of biotopes developed to promote biodiversity in urban areas, and to look for improvement steps. Specifically, artificial wetland creation techniques were categorized into living environment and living creature classification. Being living conditions for creations, habitat environment was reviewed with a focus on water and soil environments. Living creatures were classified into plants, insects, fish, and birds. The evaluation of creation techniques was done in post-construction evaluation while considering the creation of habitats for living creatures. Intervention by users, changes in living environment and living species, and relevance of creation techniques were reviewed. Key results of this study are as follows. (1) Water environment for the living environment of creatures provides a suitable environment conditions for the living of creatures through a process easing the use of piped water. Various water depths and embankment appear to have a positive impact on the living of aquatic life. In particular, embankment covered in soil naturally played an important role as a place for the activities of aquatic insects and young fish as well as the growth of aquatic plants. (2) Various aquatic and ground plants to promote insect-diversity, shallow water, and old-tree logs had contributed greatly in increasing the types and number of insects. Aquatic insects. Aquatic insects were seen much particularly in areas where aquatic plants are rich but water is shallow than any other areas. (3) A space piled with stone to provide habitats for fish was not much used. However, it was observed that fish used embankment built with natural stones and embankment using logs in areas where water is deep. In addition, it was confirmed that 1,500 fish that had been released propagated using various depths and places for birth. (4) It was analyzed that techniques (creation of island, log setting, and creation of man-made bird nests) to provide habitats and to attract birds are not serving their roles. In such a case, it is believed that species had not increased due to the smallness as well as isolated features of the area. Based on theoretical review, they are judged to be areas that are likely to be used when a greater variety of birds is introduced. It is judged that attracting and keeping more birds at the site, such spaces need to be linked systematically in the future in terms of building eco-network while ensuring an adequate living areas. (5) In the study areas, users intervened greatly. As a result, a blockage was created preventing the normal growth of plants and non-indigenous plants were introduced. In order to limit the intervention by users, setting enough buffer zones, and environment education programs were urgently required. D/H=1>Hyangkyo> houses on the river>temples>lecture halls. D/H ratio of the backside areas is as follows. D/H=1>Hyangkyo>houses on the river>lecture halls. 4. Inner garden were planted deciduous than evergreen trees with Lagerstroemia indica. Enclosed dominant trees were planted by Pinus densiflora, Querces seuata. construct GEM strain, and examined for the expression and functional stability in microcosms.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.39 no.5
• /
• pp.486-493
• /
• 2002

In this paper, we proposed the effective multispectral image compression method using CIBP(classified interband bidrectional prediction) and extended SPIHT(set partition in hierarchical trees) in wavelet domain. We determine separately feature bands that have the highest correlation with other bands in the visible range and in the infrared range of wavelengths. Feature bands are coded to remove the spatial redundancy with SPIHT in the wavelet domain. Prediction bands that have high correlation with feature bands are wavelet transformed and they are classified into one of three classes considering reflection characteristics of the baseband. For Prediction bands, CIBP is performed to reduce the spectral redundancy. for the difference bands between prediction bands and the predicted bands, They are ordered to upgrade the compression efficiency of extended SPIHT with the largest error magnitude. The arranged bands are coded to compensate the prediction error with extended SPIHT. Experiments are carried out on the multispectral images. The results show that the proposed method reconstructs higher quality images than images reconstructed by the conventional methods at the same bit rate.

• Korean Journal of Environment and Ecology
• /
• v.28 no.6
• /
• pp.725-740
• /
• 2014

To figure out the vegetation situation in the ridge of the Naknam-Jeongmaek, six intensive survey sites were selected considering environmental factors and human interferences, 132 quadrats (area $100m^2$) were installed to survey appearing species of trees and major environmental factors, and the characteristics of vegetation structures were analyzed. The surveyed plots were divided into six groups according to the analysis of classification by TWINSPAN; Quercus mongolica-Q. variabilis community, Pinus densiflora-Q. dentata community, Chamaecyparis obtusa community, Q. mongolica-P. densiflora community, P. densiflora-P. thunbergii community, P. koraiensis community, Rhododendron spp.-Lespedeza cyrtobotrya community. P. densiflora-Q. dentata community, Q. mongolica-P. densiflora community, and P. densiflora-P. thunbergii community are expected to be succeeded by deciduous oaks because the power of deciduous oaks is strong in their lower layer. C. obtusa community, P. densiflora community, and Rhododendron spp.-L.cyrtobotrya community are artificial forests that were artificially formed and are expected to be maintained in the current state for some time because the dominance value of planted species of trees is high. Most vegetations in Naknam-Jeongmaek were secondary forests or artificial forests formed for forest tree production and forestation for erosion control. In particular the top regions and hilly sections of the mountain were mostly dominated by deciduous oaks such as Q. mongolica, Q. variabilis showed some P. densiflora community competing with deciduous oaks. On the other hand, low sections and regions adjacent to the city showed severe artificial interference since exotic species such as P. thunbergiil, C. obtusa, P. koraiensis, and Rhododendron spp. were planted.

• The Korean Journal of Applied Statistics
• /
• v.28 no.5
• /
• pp.921-932
• /
• 2015

This study analyzed military personnel survivability in regards to offensive operations according to the scientific military training data of a reinforced infantry battalion. Scientific battle training was conducted at the Korea Combat Training Center (KCTC) training facility and utilized scientific military training equipment that included MILES and the main exercise control system. The training audience freely engaged an OPFOR who is an expert at tactics and weapon systems. It provides a statistical analysis of data in regards to state-of-the-art military training because the scientific battle training system saves and utilizes all training zone data for analysis and after action review as well as offers training control during the training period. The methodologies used the Cox PH modeling (which does not require parametric distribution assumptions) and decision tree modeling for survival data such as CART, GUIDE, and CTREE for richer and easier interpretation. The variables that violate the PH assumption were stratified and analyzed. Since the Cox PH model result was not easy to interpret the period of service, additional interpretation was attempted through univariate local regression. CART, GUIDE, and CTREE formed different tree models which allow for various interpretations.

• Journal of Korean Society of Forest Science
• /
• v.108 no.2
• /
• pp.189-199
• /
• 2019

The forest management paradigm has recently shifted from focusing on commercial production to focusing on ecosystem management. Accordingly, a natural seedling regeneration method that has a naturally high affinity has attracted much attention in recent years. The aim of this study was to determine the relationship among various environmental factors, lower stratum vegetation, and seedling regeneration in Pinus densiflora forests. The survey site comprised 50 sectors divided using the line transect method, and the survey data were divided into those from wet habitat (19 sites) and dry habitat (31 sites), depending on the soil humidity, and were analyzed separately to show the close relationship between soil humidity and natural seedling regeneration. As a result, the dry habitat exhibited high seedling density (157,419 trees/ha), with the main species being Quercus serrata, Zanthoxylum piperitum, Smilax china, and Pueraria lobata, while wet habitat exhibited low seedling density (57,895 trees/ha), with the main species being Stephanandra incisa, Castanea crenata, Lespedeza maximowiczii, Lysimachia barystachys, Aralia elata, and Styrax japonicus. The height and root-collar diameter under wet conditions exhibited faster growth than those under dry conditions. Height growth by the root-collar diameter in dry habitat increased faster than that in wet habitat. It was also confirmed that seedling regeneration in wet habitat exhibited a rapid growth pattern 5 years after germination. These results suggest that the seedlings begin to grow more rapidly after a period of suppression by competition with surrounding plants. Considering an ecosystem or ecological management approach, specific practices, such as bush control and vine clearing in wet habitats, should be more intensively conducted, especially at the beginning of the management operations.

• Journal of Intelligence and Information Systems
• /
• v.18 no.2
• /
• pp.29-45
• /
• 2012

Bond rating is regarded as an important event for measuring financial risk of companies and for determining the investment returns of investors. As a result, it has been a popular research topic for researchers to predict companies' credit ratings by applying statistical and machine learning techniques. The statistical techniques, including multiple regression, multiple discriminant analysis (MDA), logistic models (LOGIT), and probit analysis, have been traditionally used in bond rating. However, one major drawback is that it should be based on strict assumptions. Such strict assumptions include linearity, normality, independence among predictor variables and pre-existing functional forms relating the criterion variablesand the predictor variables. Those strict assumptions of traditional statistics have limited their application to the real world. Machine learning techniques also used in bond rating prediction models include decision trees (DT), neural networks (NN), and Support Vector Machine (SVM). Especially, SVM is recognized as a new and promising classification and regression analysis method. SVM learns a separating hyperplane that can maximize the margin between two categories. SVM is simple enough to be analyzed mathematical, and leads to high performance in practical applications. SVM implements the structuralrisk minimization principle and searches to minimize an upper bound of the generalization error. In addition, the solution of SVM may be a global optimum and thus, overfitting is unlikely to occur with SVM. In addition, SVM does not require too many data sample for training since it builds prediction models by only using some representative sample near the boundaries called support vectors. A number of experimental researches have indicated that SVM has been successfully applied in a variety of pattern recognition fields. However, there are three major drawbacks that can be potential causes for degrading SVM's performance. First, SVM is originally proposed for solving binary-class classification problems. Methods for combining SVMs for multi-class classification such as One-Against-One, One-Against-All have been proposed, but they do not improve the performance in multi-class classification problem as much as SVM for binary-class classification. Second, approximation algorithms (e.g. decomposition methods, sequential minimal optimization algorithm) could be used for effective multi-class computation to reduce computation time, but it could deteriorate classification performance. Third, the difficulty in multi-class prediction problems is in data imbalance problem that can occur when the number of instances in one class greatly outnumbers the number of instances in the other class. Such data sets often cause a default classifier to be built due to skewed boundary and thus the reduction in the classification accuracy of such a classifier. SVM ensemble learning is one of machine learning methods to cope with the above drawbacks. Ensemble learning is a method for improving the performance of classification and prediction algorithms. AdaBoost is one of the widely used ensemble learning techniques. It constructs a composite classifier by sequentially training classifiers while increasing weight on the misclassified observations through iterations. The observations that are incorrectly predicted by previous classifiers are chosen more often than examples that are correctly predicted. Thus Boosting attempts to produce new classifiers that are better able to predict examples for which the current ensemble's performance is poor. In this way, it can reinforce the training of the misclassified observations of the minority class. This paper proposes a multiclass Geometric Mean-based Boosting (MGM-Boost) to resolve multiclass prediction problem. Since MGM-Boost introduces the notion of geometric mean into AdaBoost, it can perform learning process considering the geometric mean-based accuracy and errors of multiclass. This study applies MGM-Boost to the real-world bond rating case for Korean companies to examine the feasibility of MGM-Boost. 10-fold cross validations for threetimes with different random seeds are performed in order to ensure that the comparison among three different classifiers does not happen by chance. For each of 10-fold cross validation, the entire data set is first partitioned into tenequal-sized sets, and then each set is in turn used as the test set while the classifier trains on the other nine sets. That is, cross-validated folds have been tested independently of each algorithm. Through these steps, we have obtained the results for classifiers on each of the 30 experiments. In the comparison of arithmetic mean-based prediction accuracy between individual classifiers, MGM-Boost (52.95%) shows higher prediction accuracy than both AdaBoost (51.69%) and SVM (49.47%). MGM-Boost (28.12%) also shows the higher prediction accuracy than AdaBoost (24.65%) and SVM (15.42%)in terms of geometric mean-based prediction accuracy. T-test is used to examine whether the performance of each classifiers for 30 folds is significantly different. The results indicate that performance of MGM-Boost is significantly different from AdaBoost and SVM classifiers at 1% level. These results mean that MGM-Boost can provide robust and stable solutions to multi-classproblems such as bond rating.

• Journal of Korean Society of Forest Science
• /
• v.14 no.1
• /
• pp.1-20
• /
• 1972

The trend and achievements of forest genetics research in abroad were investigated through observation tours and reference work and following facts were found to be important aspects which should be adopted in the forest genetics research program in Korea. Because of world wide recognization on the urgency of taking a measure to reserve some areas of the representative forest type on the globe before the extingtion of such forest type as the results of continuous exploitations of the natural forests to meet the timber demand all over the world, it is urgently needed to take a measure to reserve certain areas of natural stand of Pinus koraiensis, Pinus parviflora, Pinus densiflora f. erectra, Abies koreana, Quercus sp., Populus sp., etc. as gene pool to be used for the future program of forest tree improvement. And the genetic studies of those natural forest of economic tree species are also to be performed. 1. Increase of the number of selected tree for breeding purpose. Because of the fact that the number of plus tree at present is too small to carry out selection program for tree improvement, particularly for the formation of source population for recurrent selection of parent trees of the 2nd generation seed orchard it is to be strongly emphasized to increase the number of plus tree by alleviating selection criteria in order to enlarge the population size of plus trees to make the selection program more efficient. 2. Progeny testing More stress should be placed on carrying out progeny testing of selected trees with open pollinated seeds. And particular efforts are to be made for conducting studies on adult/juvenile correlation of important traits with a view to enable to predict adult performances with some traits revealed in juvenile age thus to save time for progeny testing. 3. Genotype-environment interaction Studies on genotype and environment interaction should be conducted in order to elucidate whether the plus trees selected on the good site express their superiority on the poor site or not and how the environment affect the genotype. And the justification of present classification of seed distribution area should be examined. 4. Seed orchard of broad leaf tree species. Due to the difficulty of accurate comparison of growth rate of neighbouring trees of broad leaf tree species in natural stand, it is recommended that for the improvement of broad leaf trees a seedling seed orchard is to be made by roguing the progeny test plantation planted densely with control pollinated seedlings of selected trees. 5. Breeding for insect resistant varieties. In the light of the fact that the resistant characteristics against insect such as pine gall midge (Thiecodiplosis japonensis U. et I.) and pine bark beetle (Myelophilus pinipera L.) are highly correlated with the amount and quality of resin which are known as gene controlled characteristics, breeding for insect resistance should be carried out. 6. Breeding for timber properties. With the tree species for pulp wood in particular, emphasis should be placed upon breeding for high specific gravity of timber. 7. Introduction of Cryptomeria and Japanese Cypress In the light of the fact that the major clones of Cryptomeria are originated from Yoshino source and are being planted up to considerably north and high elevation in Japan, those species should be examined on their cold resistance in Korea by planting them in further northern part of the country.