• Communications for Statistical Applications and Methods
• /
• v.24 no.6
• /
• pp.543-559
• /
• 2017

Tree-based regression and classification ensembles form a standard part of the data-science toolkit. Many commonly used methods take an algorithmic view, proposing greedy methods for constructing decision trees; examples include the classification and regression trees algorithm, boosted decision trees, and random forests. Recent history has seen a surge of interest in Bayesian techniques for constructing decision tree ensembles, with these methods frequently outperforming their algorithmic counterparts. The goal of this article is to survey the landscape surrounding Bayesian decision tree methods, and to discuss recent modeling and computational developments. We provide connections between Bayesian tree-based methods and existing machine learning techniques, and outline several recent theoretical developments establishing frequentist consistency and rates of convergence for the posterior distribution. The methodology we present is applicable for a wide variety of statistical tasks including regression, classification, modeling of count data, and many others. We illustrate the methodology on both simulated and real datasets.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.30 no.4A
• /
• pp.241-249
• /
• 2005

ATM based MPLS(Multiprotocol Label Switching) is discussed for its provisioning QOS commitment capabilities, Traffic engineering and smooth migration for BcN in Korea. At this time, due to the comprehensive nature of ATM protocol, ATM has been adapted as the backbone system for carrying Internet traffic[1,2,3,4]. This paper presents preventive congestion control mechanisms for detecting HTR(Hard-To-Reach) LSP(Label Switched Path) in ATM based MPLS systems. In particular, we have introduced a HTR LSP detection method using network signaling information in an ATM layer. MPLS related studies can cover LSP failures in a physical layer fault, it can not impact network congestion status. Here we will present the research results for introducing HTR LSP detection methods and control mechanisms and this mechanism can be implementing as SOC for high speed processing a packet header. We concluded that it showed faster congestion avoidance abilities with a more reduced system load and maximized the efficiency of network resources by restricting ineffective machine attempts.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2004.05a
• /
• pp.1037-1040
• /
• 2004

Sound and vibrations enter every aspect of our daily lives. Some, like music and verbal communications, soothe and relax us; others, such as the noise and vibration from traffic or machinery, can threaten our well being. The Rion Co., Ltd. was eslablished in 1944 as an affiliate of physics and acoustic pioneer, Kobayashi Institute of Physical Research. The RION lineup affords a full line of easy-to handle measuring, recording, and analyzing equipment with innumerable applications for monitoring and maintaining a comfortable environment. RION's concern takes in all manner of sound and vibration. Our activities extend from measuring reverberation time in concert halls and testing the soundproofing characteristics of housing materials to the control and maintenance of machine noise and vibration levels.

• Proceedings of the KIPE Conference
• /
• 2008.06a
• /
• pp.592-594
• /
• 2008

The powers used in the robot drive motor system are mostly DC sources such as batteries. Even AC powers in some systems are generated from DC sources by the inverter. It is can be forecasted that the DC-link communications will be widely used invarious industrial application. In this paper a novel BLDC motor drive system by using DC-link communications is proposed. The characteristic of this system is the communication only needs 2 DC lines. There are not additional lines to translate the reference signals and the reliability of the system is enhanced especially in some badly circumstance. The number of lines can be least when applied in the multi motor control system and the slip ring design can be simplified when applied as rotation machine. The reasonableness of proposed motor system is clarified by the PSIM simulation and the hardware prototype.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.4C
• /
• pp.425-432
• /
• 2007

In this paper, we deals with automatic modulation classification capable of classifying incident signals without a priori information. The 7 key features which have good properties of sensitive with modulation types and insensitive with SNR variation are selected. The numerical simulations for classifying 9 modulation types using the these features are performed. The numerical simulations of the 4 types of modulation classifiers are performed the investigation of classification accuracy and execution time to implement the fast modulation classifier in software radio. The simulation result indicated that the execution time of DTC was best and SVC and MDC showed good classification performance. The prototype was implemented with DTC type. With the result of field trials, we confirmed the performance in the prototype was agreed with the numerical simulation result of DTC.

• ETRI Journal
• /
• v.30 no.1
• /
• pp.113-128
• /
• 2008

Turbo codes are extensively used in current communications standards and have a promising outlook for future generations. The advantages of software defined radio, especially dynamic reconfiguration, make it very attractive in this multi-standard scenario. However, the complex and power consuming implementation of the maximum a posteriori (MAP) algorithm, employed by turbo decoders, sets hurdles to this goal. This work introduces an ASIP architecture for the MAP algorithm, based on a dual-clustered VLIW processor. It displays the good performance of application specific designs along with the versatility of processors, which makes it compliant with leading edge standards. The machine deals with multi-operand instructions in an innovative way, the fetching and assertion of data is serialized and the addressing is automatized and transparent for the programmer. The performance-area trade-off of the proposed architecture achieves a throughput of 8 cycles per symbol with very low power dissipation.

• Communications for Statistical Applications and Methods
• /
• v.15 no.1
• /
• pp.43-50
• /
• 2008

Multinomial logistic regression is a well known multiclass classification method in the field of statistical learning. More recently, the development of sparse multinomial logistic regression model has found application in microarray classification, where explicit identification of the most informative observations is of value. In this paper, we propose a sparse multinomial kernel logistic regression model, in which the sparsity arises from the use of a Laplacian prior and a fast exact algorithm is derived by employing a bound optimization approach. Experimental results are then presented to indicate the performance of the proposed procedure.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.13 no.3
• /
• pp.231-238
• /
• 1988

A new approach to the reachability analysis of communication protocols is presented using a modified maximal progress state exploation for two communicating finite state machines. The task of generating all reachable states is divided into two independent subtasks. In each subtask, only the states which are reachable by forcing modified maximal progress sequence for one machine are generated. Modified maximal progress state exploration saves space and time over maximal progress state exploration.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.11 no.2
• /
• pp.1-6
• /
• 2012

Recent ultra-precision machining systems have nano-scale resolution, and can machine various shapes of complex structures using five-axis driven modules. These systems are also multi-functional, which can perform various processes such as planing, milling, turning et al. in one system. Micro machining technology using these systems is being developed for machining fine patterns, hybrid patterns and high aspect-ratio patterns on large-area molds with high productivity. These technology is and will be applied continuously to the fields of optics, display, energy, bio, communications and et al. Domestic and foreign trends of micro machining technologies for flat molds were investigated in this study. Especially, we focused on the types and the characteristics of ultra-precision machining systems and application fields of micro patterns machined by the machining system.

• Asia Marketing Journal
• /
• v.14 no.4
• /
• pp.77-94
• /
• 2013

Really new products (RNPs) provide novel benefits yet many consumers are reluctant to accept these highly innovative new products. Previous literature has shown that mental simulation is an effective method for enhancing the evaluation of RNPs. However, Castano et al. (2008) and Zhao, Hoeffler, and Zauberman (2011) demonstrate conflicting results as to which type of mental simulation (i.e., process versus outcome) is more effective for the enhancement of RNP evaluation. The authors try to reconcile these results by incorporating a moderating variable which is personal need for structure (PNS). PNS is an individual difference variable that taps the differences in people's propensity to cognitively structure and simplify their environment (Neuberg and Newsom 1993). From the analysis of the previous two works, the authors point out that consumers' susceptibility to uncertainty may contribute to the different results, and suggest that this susceptibility is dependent on consumers' PNS. To test the hypotheses established, an experiment was conducted. Waterless washing machine was presented as a RNP and PNS was measured by using the 12-item PNS Scale (Thompson et al. 2001). The results of the study show that for high-PNS consumers, process simulation is more effective than outcome simulation for enhancing the evaluation of a RNP, whereas for low-PNS consumers, outcome simulation is more effective than process simulation. This research contributes to the mental simulation and new product literature by suggesting and verifying that PNS moderates the effects of process versus outcome simulations for enhancing the evaluation of RNPs. This research provides important managerial implications for marketing managers of RNPs, indicating that they should take account of the target consumers' PNS in planning marketing communications. Specifically, when targeting high-PNS consumers, marketing communications that encourage process simulation may be more effective than those that encourage outcome simulation. In contrast, when targeting low-PNS consumers, marketing communications that encourage outcome simulation may be more effective than those that encourage process simulation.