• Communications for Statistical Applications and Methods
• /
• v.13 no.2
• /
• pp.449-466
• /
• 2006

We propose to use variable selection methods based on penalized regression for pruning decision tree ensembles. Pruning methods based on LASSO and SCAD are compared with the cluster pruning method. Comparative studies are performed on some artificial datasets and real datasets. According to the results of comparative studies, the proposed methods based on penalized regression reduce the size of boosting ensembles without decreasing accuracy significantly and have better performance than the cluster pruning method. In terms of classification noise, the proposed pruning methods can mitigate the weakness of AdaBoost to some degree.

• Korean Journal of Soil Science and Fertilizer
• /
• v.49 no.5
• /
• pp.577-583
• /
• 2016

Small-sized persimmons produced by high crop load are better accepted in the export markets. However, maintaining high crop load frequently results in weakness of tree vigor, deterioration of fruit quality, and increase of the risks for alternate bearing. This experiment was conducted to determine the combined effects of fertilization rate and leaf-fruit (L/F) ratio on container-grown 3-year-old 'Fuyu' persimmon trees. Application of 3.6-g N, 2.1-g $P_2O_5$, 2.7-g $K_2O$, 2.7-g CaO, and 0.6-g MgO was for the control fertilization rate (CF) and that of a 3-fold CF was for the high fertilization rate (HF). Commercial fertilizers were surface-applied to a container on July 6, July 17, and August 10 in three equal aliquots. Single tree for each fertilization rate was assigned for 12 L/F ratios (5, 6.3, 7.7, 9, 10.4, 13, 15.5, 18, 21, 24, 27, and 33) mostly by fruit thinning or rarely by defoliation on July 1. HF did not affect the yield, weight and soluble solids of the fruits but decreased skin color. As L/F ratio increased, yield decreased but average weight, skin color, and soluble solids of fruits increased. With HF, N and K concentrations in leaves, fruits, and shoots increased to some extent but soluble sugars in dormant shoots decreased. Many shoots were cold-injured with low L/F ratio especially at the HF. HF did not increase number of flower buds the next spring either on a shoot or on a tree basis but increased shoot length, compared with the CF. Increasing L/F ratio markedly increased number of flower buds and shoot growth the following year at both fertilization rates. Therefore, an appropriate combination of fertilization rate and L/F ratio should be necessary to maintain stable fruit production and tree vigor at high crop load.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.04a
• /
• pp.576-579
• /
• 2011

Wireless sensor network are widely all over different fields. Because of its distinguished characteristics, we must take account of the factor of energy consumed when designing routing protocol. Wireless sensor networks consist of small battery powered devices with limited energy resources. Once deployed, the small sensor nodes are usually inaccessible to the user, and thus replacement of the energy source is not feasible. Hence, energy efficiency is a key design issue that needs to be enhanced in order to improve the life span of the network. In BCDCP, all sensors sends data from the CH (Cluster Head) and then to the BS (Base Station). BCDCP works well in a smallscale network however is not preferred in a large scale network since it uses much energy for long distance wireless communication. TBRP can be used for large scale network, but it weakness lies on the fact that the nodedry out of energy easily since it uses multi-hops transmission data to the Base Station. Here, we proposed a routing protocol. A Cluster Based Energy Efficient Tree Routing Protocol (CETRP) in Wireless Sensor Networks (WSNs) to prolong network life time through the balanced energy consumption. CETRP selects Cluster Head of cluster tree shape and uses maximum two hops data transmission to the Cluster Head in every level. We show CETRP outperforms BCDCP and TBRP with several experiments.

• Journal of Korea Multimedia Society
• /
• v.13 no.1
• /
• pp.133-142
• /
• 2010

As the ubiquitous technology has penetrated into almost every aspect of modern life, the research of the security technology to solve the weakness of security in the ubiquitous environment is received much attention. Because, however, today's security systems are usually based on the fixed rules, many security systems can not handle diverse situations in the ubiquitous environment appropriately. Although many existing researches on context aware security service are based on ACL (Access Control List) or RBAC (Role Based Access Control), they have an overhead in the management of security policy and can not manipulate unexpected situations. Therefore, in this paper, we propose a context-aware security service providing multiple authentications and authorization from a security level which is decided dynamically in a context-aware environment using FCM (Fuzzy C-Means) clustering algorithm and Fuzzy Decision Tree. We show proposed model can solve typical conflict problems of RBAC system due to the fixed rules and improve overhead problem in the security policy management. We expect to apply the proposed model to the various applications using contextual information of the user such as healthcare system, rescue systems, and so on.

• Journal of Korea Spatial Information System Society
• /
• v.12 no.1
• /
• pp.49-56
• /
• 2010

Many researches of frequent pattern mining technique for detecting unknown patterns on spatial data have studied actively. Existing data structures have classified into tree-structure and array-structure, and those structures show the weakness of performance on dense or sparse data. Since spatial data have obtained the characteristics of dense and sparse patterns, it is important for us to mine quickly dense and sparse patterns using only single algorithm. In this paper, we propose novel data structure as compressed patricia frequent pattern tree and frequent pattern mining algorithm based on proposed data structure which can detect frequent patterns quickly in terms of both dense and sparse frequent patterns mining. In our experimental result, proposed algorithm proves about 10 times faster than existing FP-Growth algorithm on both dense and sparse data.

• The KIPS Transactions:PartD
• /
• v.11D no.7 s.96
• /
• pp.1417-1426
• /
• 2004

As Ubiquitous commerce is coming personalization service is getting interested. And also the recommendation method which offers useful information to customer becomes more important. However, most of them depend on specific method and are restricted to the E-commerce. For applying these recommendation methods into U-commerce, first it is necessary that the extended context modeling and systematic connection of the methods to complement strength and weakness of recommendation methods in each commercial transaction. Therefore, we propose a mod-eling technique of context information related to personal activation in commercial transaction and show incremental preference analysis method, using preference tree which is closely connected to recommendation method in each step. And also, we use an XML indexing technique to effi-ciently extract the recommendation information from a preference tree.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.2 no.1
• /
• pp.58-65
• /
• 1999

The purpose of this study was to evaluate the effect of rehabilitation measures on the fired woodland of Chesuk-bong in Mt. Chiri through the vegetation structure and comparison of growth pattern between natural seedlings and transplanted Abies koreana seedlings. It was also discussed the rate of standing and fallen trees of dead conifers in the management strategy for vegetation regeneration and protection of sub-alpine area. In this fired woodland, shrub layers consisting of Weigela subsessilis, Sambucus williamsii var. coreana, Rhododendron schlippenbachii, R. mucronulatum, Tripterygium regelii, and etc. were gradually expanded, and tree species such as Betula ermani, Sorbus commixta, Acer pseudosieboldianum, and Malus baccata var. mandshurica were also regenerated. The growth of height and crown width of Abies koreana seedlings transplanted for vegetation regeneration were a little lower than those of natural seedlings. The seedlings of A. koreana transplanted in this fired woodland grew about 50.6% normally, but the others had multi-branched or growth stopped by death of terminal shoot. Because the number of dead conifers by fire tend to be gradually increased as time passed, it can occurs to the soil erosion and landslide by weakness of the cohesion and resistance of soil. Therefore, it is consider that rehabilitation measures projects of the fired woodland in subalpine area may be more prevention a natural disaster like soil erosion, flood and landslide.

• Pediatric Gastroenterology, Hepatology & Nutrition
• /
• v.5 no.2
• /
• pp.186-191
• /
• 2002

Choledochal cyst is a congenital anomaly with classic triad of abdominal pain, jaundice and right upper abdominal mass. Bile peritonitis caused by cyst rupture is relatively not rare in infancy. The mechanism of rupture must be epithelial irritation of the biliary tract by refluxed pancreatic juice caused by pancreatico-biliary malunion associated with mural immaturity in infancy, rather than an abnormal rise in ductal pressure or congenital mural weakness at a certain point. We experienced a case of bile peritonitis caused by spontanenous rupture of choledochal cyst in a 10-month-old girl presented with abdominal distension, persistent fever, diarrhea, irritability and intractable ascites. She was presumed as having bile peritonitis by bile colored ascitic fluid with elevated bilirubin level and diagnosis was made by $^{99m}Tc$ DISIDA hepatobiliary scan showing extrahepatic biliary leak. The perforated cyst was surgically removed and the biliary tree was reconstructed with a Roux-en-Y hepaticojejunostomy.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.3
• /
• pp.263-274
• /
• 2010

Mutual exclusion that applied on existing systems was designed for static distributed systems. but now computing environments are going to mobile computing environments. Therefore a mutual exclusion algorithm in static distributed environments should be designed for new computing environments. So this paper proposes a mobile mutual exclusion algorithm to support the mutual exclusion of shared resources in mobile computer environments. Mobile computing resources as wireless hosts cause new issues because of their mobility and weakness and made mutual exclusion problem more complex than stationary distributed environments. So we proposed a new mobile token mutual exclusion algorithm with deadlock-free and starvation-free in mobile computing environments based on spanning tree topology and extend for mobile computing environments. The proposed algorithm minimizes message complexity in case of free hopping in cellular networks.

• The Korean Journal of Applied Statistics
• /
• v.19 no.1
• /
• pp.149-161
• /
• 2006

The most important component in decision tree algorithm is the rule for split variable selection. Many earlier algorithms such as CART and C4.5 use greedy search algorithm for variable selection. Recently, many methods were developed to cope with the weakness of greedy search algorithm. Most algorithms have different selection criteria depending on the type of variables: continuous or nominal. However, ordinal type variables are usually treated as continuous ones. This approach did not cause any trouble for the methods using greedy search algorithm. However, it may cause problems for the newer algorithms because they use statistical methods valid for continuous or nominal types only. In this paper, we propose a ordinal variable selection method that uses Cramer-von Mises testing procedure. We performed comparisons among CART, C4.5, QUEST, CRUISE, and the new method. It was shown that the new method has a good variable selection power for ordinal type variables.