• Journal of Intelligence and Information Systems
• /
• v.7 no.1
• /
• pp.17-25
• /
• 2001

Tree model is the most popular classification algorithm in data mining due to easy interpretation of the result. In CART(Breiman et al., 1984) and C4.5(Quinlan, 1993) which are representative of tree algorithms, the split fur classification proceeds to attain the homogeneous terminal nodes with respect to the composition of levels in target variable. But, fur instance, in the chum prediction modeling fur CRM(Customer Relationship management), the rate of churn is generally very low although we are interested in mining the churners. Thus it is difficult to get accurate prediction modes using tree model based on the traditional split rule, such as mini or deviance. Buja and Lee(1999) introduced a new split rule, one-sided purity for classifying minor interesting group. In this paper, we compared one-sided purity with traditional split rule, deviance analyzing churning vs. non-churning data of ISP company. Also reviewing the result of tree model based on one-sided purity with some simulated data, we discussed problems and researchable topics.

• The KIPS Transactions:PartD
• /
• v.11D no.4
• /
• pp.809-822
• /
• 2004

In this paper, we propose a concurrency control algorithm based on R-tree for spatial database management system. The previous proposed algorithms can't prevent problem that search operation is to be blocking during update operations. In case of multidimensional indexes like R-tree, locking of update operations may be locked to several nodes, and splitting of nodes have to lock a splitting node for a long time. Therefore search operations have to waiting a long time until update operations unlock. In this paper we propose new algorithms for lock-free search operation. First, we develop a new technique using a linked-list technique on the node. The linked-list enable lock-free search when search operations search a node. Next, we propose a new technique using a version technique. The version technique enable lock-free search on the node that update operations is to be splitting.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.2
• /
• pp.127-130
• /
• 2005

Internal cities such as gray level been enclosed to building forest are paying a lot of efforts and expenses to change to green city that park and street tree get put together. By the example, 'GREEN CITY of PUSAN 21' progress to decorate army facilities like the park, and to plant street trees in several places of city plan in Pusan. And urban environment that big cities of advanced nation are agreeable is making in the large park and road street trees at several places in downtown. Because price of tree for the park is very expensive, tree management system was all-important. In this paper, Motes deliver the sensor information in each tree through radio sensor network by server side. This information can use in state grasping of tree, harmful insects courtesy call etc and this system design was suggested to inform to mode of life administration scholars.

• Journal of the Korean Society of Environmental Restoration Technology
• /
• v.7 no.2
• /
• pp.36-45
• /
• 2004

This study was conducted to formulate management guidelines for monumental old trees in Korea through analysis of growing environments. A total of 20 old trees designated as natural monuments in Seoul, Incheon, and Gyeonggi provinces were surveyed for biological characteristics, surrounding environments, root collar conditions, tree health, and soil characteristics. Relationships among root collar conditions, tree health, and soil characteristics were analyzed by correlation. The old solitary trees designated as natural monuments included Pinus bungeana(4 trees), Juniperus chinensis(3 trees), Ginkgo biloba(3 trees), Poncirus trifoliata(2 trees), Actinidia arguta, Wisteria floribunda, Thuja orientalis, Quercus variabilis, Sophora japonica, Fraxinus rhynchophylla, Zelkova serrata, and Pinus densiflora. The tree height ranged from 3.56 to 67m, and root collar diameter ranged from 1.01 to 15.2m. The monumental old trees were growing on the various sites ranging from gardens, historical sites, open agricultural fields, mountain hills, to near the ocean beaches and streams. The coverage of bald land ranged from 50 to 100%, and depth of filled soil around the root collar ranged from 0 to 50cm. Tree health was expressed as the amount of branch dieback, cavity development, detachment of cambial tissue, infliction by diseases and insects. The branch dieback ranged from 5 to 20%, cavity development ranged from 10 to 100$cm^3$, detachment of cambial tissue ranged from 5 to 45%, and infliction by diseases and insects ranged from 5 to 20%. Soil pH ranged from 5.9 to 8.3, organic matter contents from 12 to 56%, phosphorus contents from 104 to 618ppm, while soil compaction ranged from 7 to 28mm. Results of correlation analysis showed that coverage of bald land was the most serious factor to deteriorate the cavity development and detachment of cambial tissue. In addition, chemical properties of soils seemed to be related to the health of the trees.

• Journal of Digital Convergence
• /
• v.14 no.8
• /
• pp.415-421
• /
• 2016

A behavior tree is to express the behavior of artificial intelligence. The behavior tree has a characteristic that is easy to change state transitions than FSM(Finite State Machine), see the progress of the action. For these reasons, the behavior tree is widely used in more than FSM. This paper is to analyze the advantages and disadvantages on behavior trees of game engines, proposes the improved behavior tree based on analyzed them. To achieve this, in this paper, first, examines the role of node and the behavior tree structure of the unity engine, unreal engine. Second, discusses the advantages and disadvantages based on it. Third, proposes the behavior tree to improve the disadvantages of behavior tree of unity engine and unreal engine, depth of behavior tree and search time required to select the execution node. This paper can help developers using the tree to develop the game.

• Korean Journal of Environmental Agriculture
• /
• v.42 no.2
• /
• pp.104-111
• /
• 2023

As apple trees mature, it is important to maintain good light distribution within the canopy to produce marketable fruits. Tree thinning is the selective removal of a proportion of trees growing in the orchard to provide more growing space and a good light environment for the remaining trees. This study was conducted over 3 years (14-16 years after planting) to investigate the influence of tree thinning on vegetative growth, yield, fruit quality, and blooming in the 14 years old slender spindle 'Fuji'/M.9 apple trees planted with the tree space of 3.2×1.2 m. The trees were placed in a control group (no thinning; 260 trees per 10 a) or a tree thinning group (thinned 50% of the control; 130 trees per 10 a). The tree thinning successfully improved light penetration, yield per tree, fruit red color, and yield efficiency for 3 years, and the tree thinning controlled the occurrence of biennial bearing. However, tree thinning significantly decreased accumulated yield per 10 a compared with the control. The vegetative growth, yield per tree, soluble solid contents, and blooming were not clear by the occurrence of biennial bearing in the control. These results indicated that tree thinning was a good method for improving light penetration and preventing biennial bearing in the old 'Fuji'/M.9 high-density apple orchards.

• Journal of Platform Technology
• /
• v.6 no.3
• /
• pp.47-54
• /
• 2018

Predictive analysis in the hardware industry can be performed at an appropriate point in time to prevent failure of production facilities and reduce management costs. This helps to perform more efficient and scientific maintenance through automation of failure analysis. Among them, predictive management aims to prevent the occurrence of anomalous state by identifying and improving the abnormal state based on the gathering, analysis, and scientific data management of facilities using information technology and constructing prediction model do. In this study, we made a fault tree through the Relex tool and analyzed the error code of the hardware to study the safety.

• Journal of Forest and Environmental Science
• /
• v.36 no.1
• /
• pp.7-16
• /
• 2020

Tree size distribution modelling is an integral part of forest management. Most distribution yield systems rely on some flexible probability models. In this study, a simple finite mixture of two components two-parameter Weibull distribution was compared with complex four-parameter distributions in terms of their fitness to predict tree size distribution of teak (Tectona grandis Linn f) plantations. Also, a system of equation was developed using Seemingly Unrelated Regression wherein the size distributions of the stand were predicted. Generalized beta, Johnson's SB, Logit-Logistic and generalized Weibull distributions were the four-parameter distributions considered. The Kolmogorov-Smirnov test and negative log-likelihood value were used to assess the distributions. The results show that the simple finite mixture outperformed the four-parameter distributions especially in stands that are bimodal and heavily skewed. Twelve models were developed in the system of equation-one for predicting mean diameter, seven for predicting percentiles and four for predicting the parameters of the finite mixture distribution. Predictions from the system of equation are reasonable and compare well with observed distributions of the stand. This simplified mixture would allow for wider application in distribution modelling and can also be integrated as component model in stand density management diagram.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.4
• /
• pp.73-82
• /
• 2012

Currently, R&D investment of government is increased dramatically. However, the budget of the government is different depending on the size of ministry and priorities, and then it is difficult to obtain consensus on the budget. They did not establish decision support systems to evaluate and execute R&D budget. In this paper, we analyze factors affecting research funds by linear regression and decision tree analysis in order to increase investment efficiency in national research project. Moreover, we suggested strategies that budget is estimated reasonably.

• ETRI Journal
• /
• v.27 no.4
• /
• pp.355-366
• /
• 2005

Automatic discovery of physical topology plays a crucial role in enhancing the manageability of modern metro Ethernet networks. Despite the importance of the problem, earlier research and commercial network management tools have typically concentrated on either discovering logical topology, or proprietary solutions targeting specific product families. Recent works have demonstrated that network topology can be determined using the standard simple network management protocol (SNMP) management information base (MIB), but these algorithms depend on address forwarding table (AFT) entries and can find only spanning tree paths in an Ethernet mesh network. A previous work by Breibart et al. requires that AFT entries be complete; however, that can be a risky assumption in a realistic Ethernet mesh network. In this paper, we have proposed a new physical topology discovery algorithm which works without complete knowledge of AFT entries. Our algorithm can discover a complete physical topology including inactive interfaces eliminated by the spanning tree protocol in metro Ethernet networks. The effectiveness of the algorithm is demonstrated by implementation.