• The Korean Journal of Applied Statistics
• /
• v.36 no.5
• /
• pp.381-397
• /
• 2023

The most representative design used in clinical trials is randomization, which is used to accurately estimate the treatment effect. However, comparison between the treatment group and the control group in an observational study without randomization is biased due to various unadjusted differences, such as characteristics between patients. Propensity score weighting is a widely used method to address these problems and to minimize bias by adjusting those confounding and assess treatment effects. Inverse probability weighting, the most popular method, assigns weights that are proportional to the inverse of the conditional probability of receiving a specific treatment assignment, given observed covariates. However, this method is often suffered by extreme propensity scores, resulting in biased estimates and excessive variance. Several alternative methods including trimming, overlap weights, and matching weights have been proposed to mitigate these issues. In this paper, we conduct a simulation study to compare performance of various propensity score weighting methods under diverse situation, such as limited overlap, misspecified propensity score, and treatment contrary to prediction. From the simulation results overlap weights and matching weights consistently outperform inverse probability weighting and trimming in terms of bias, root mean squared error and coverage probability.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.9 no.3
• /
• pp.31-36
• /
• 2009

Bayesian networks are known as the best tools to express and predict the domain knowledge with uncertain environments. However, bayesian learning could be too difficult to do effective and reliable searching. To solve the problems of overtime demand, the nodes should be arranged orderly, so that effective structural learning can be possible. This paper suggests the classification learning model to reduce the errors in the independent condition, in which a lot of variables exist and data can increase the reliability by calculating the each entropy of probabilities depending on each circumstances. Also efficient learning models are suggested to decide the order of nodes, that has lowest entropy by calculating the numerical values of entropy of each node in K2 algorithm. Consequently the model of the most suitably settled Bayesian networks could be constructed as quickly as possible.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.11 no.11
• /
• pp.4174-4181
• /
• 2010

This paper presents an adaptive scheduling algorithm for manufacturing processes with non-stationary rework probabilities. The adaptive scheduling scheme named by hybrid Q-learning algorithm is proposed in this paper making use of the non-stationary rework probability and coupling with artificial neural networks. The proposed algorithm is measured by mean tardiness and the extensive computational results show that the presented algorithm gives very efficient schedules superior to the existing dispatching algorithms.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.936-940
• /
• 2007

본 논문은 무선 센서 네트워크의 에너지 효율적인 운영을 위해 무선 센서 네트워크 환경에 적합한 클러스터 헤드 선출 메커니즘을 제안한다. LEACH 와 같은 기존의 확률 모델 기반의 클러스터 헤드 선출 메커니즘들은 각 라운드마다 클러스터 헤드로 선출될 확률과 라운드 횟수 등을 바탕으로 클러스터 헤드를 선출한다. 그러나 이와 같은 방법은 각 노드의 상황을 고려하지 않아 네트워크의 수명을 단축시킬 수 있다. 이러한 문제점을 해결하기 위해서는 각 센서 노드의 에너지 및 노드 분포 상황을 고려하여 클러스터 헤드를 선출해야 한다. 하지만 실제 무선 센서 네트워크 환경에서는 클러스터 헤드 선출을 위해 정확한 정보를 수집하고 이를 계산하는데 있어 큰 오버헤드가 발생하는 문제점이 있다. 이에 본 논문에서는 정보 수집 및 계산에 있어서 오버헤드를 줄이고 네트워크의 수명을 극대화하기 위하여 퍼지 논리를 이용한 퍼지 논리 기반의 클러스터 헤드 선출 메커니즘을 제안한다. Matlab 을 통한 시뮬레이션 결과 LEACH 에 비해 퍼지 논리 기반의 클러스터 헤드 선출 메커니즘을 이용했을 경우 네트워크 수명이 약 16.3% 향상되었다.

• Korean Journal of Culture and Social Issue
• /
• v.21 no.2
• /
• pp.177-203
• /
• 2015

The aims of this study were 1) to compare irrational gambling beliefs of gamblers and non-gamblers, 2) to investigate the role of cognitive error on winning probability thinking error, and 3) to examine the relationship between attributional bias and gambling behavior. A total of 248 subjects were recruited for this study. All subjects were classified into non-gamblers, social gamblers and pathological gamblers, and administered self-report questionnaires to measure irrational gambling beliefs, the probability inference error, the attriburional style, and the attributional bias. A pathological gambler group scored highest on irrational gambling beliefs, especially the overestimation of self-ability factor, and a social gambler group and a non-gambler group follow. All three groups scored higher on the magnification of gambling skills than the mean (4.0) of the scale. Pathological gamblers and social gamblers scored higher on the probability thinking error than non-gamblers. Pathological gamblers displayed higher external attribution, lower internal attribution in their daily life events and higher internal attribution in failure situation than social gamblers and non-gamblers. The results indicate that cognitive errors would be a factor that differentiates pathological gamblers from social gamblers and non-gamblers. In predicting gambling behaviors, overestimation of self-ability of irrational gambling beliefs, internal attribution in failure situation, external attribution in daily live event, and probability thinking error were identified as significant factors. It is concluded that a public education about common cognitive bias featured in gamblers might be important in prevention of pathological gambling behaviors.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04a
• /
• pp.395-397
• /
• 2003

서비스거부공격(DoS)은 공격의 목표가 되는 호스트에 악의적인 패킷을 단지 하나만 보냄으로써 공격이 이루어 질 수 없다. 그렇기 때문에 침입탐지시스템에서는 패킷에 대한 일련의 순차성을 알아냄으로써 연속적인 패킷을 보내는 DoS 공격으로서의 침입을 탐지해 내는 것이 가능하다. 본 연구에서는 네트워크 패킷에 대한 송신지의 주소와 서비스, 목적지의 주소와 서비스를 이용하여 이벤트를 정의 하였으며. 이렇게 정의된 이벤트를 이용하여 정상적인 상황에서의 이벤트에 대한 순차의존성과 DoS 공격 상황에서의 이벤트에 대한 순차의존성을 알아내었다. 이벤트에 대한 순차의존성을 측정하기 위해서 정상 이벤트가 발생하는 확률을 이용하였으며. 더 나아가서는 정보이론 분야에서의 조건부 엔트로피의 사용도 제안하였다.

• Proceedings of the Korea Contents Association Conference
• /
• 2017.05a
• /
• pp.383-384
• /
• 2017

재난 재해 조기 예 경보시스템은 산불, 홍수, 지진, 화재, 붕괴, 가스 누출 등 다양한 재난 재해가 발생할 확률이 있거나 발생하였을 경우 다양한 매체를 이용하여 신속하게 이와 관련된 정보를 전파하는 시스템을 의미한다. 재난 재해 조기 예 경보시스템은 기후 변화와 현대 사회의 발전의 고도화로 각종 다양한 재난 재해의 발생이 매년 급증하고 있어 이에 대해 국민의 생명과 재산을 보호하기 위해 필요성이 날로 증가하고 있으며, 이와 관련된 재난통신, 단말기, 시스템의 시장규모는 지속적으로 빠르게 성장하고 있다. 최근 들어 IoT를 이용한 다양한 센서 디바이스를 활용하여 재난 재해 상황, 기상상황, 전염병 상황정보 등을 광범위하게 감지하고 분석함으로써 미래 적용 가능성이 상당히 높게 나타나고 있다.

• Environmental and Resource Economics Review
• /
• v.13 no.4
• /
• pp.735-761
• /
• 2004

As a possible alternative to Traditional Discounted Cash Flow Method, "Option Pricing Model" has drawn academic attentions for the last a few decades. However, it has failed to replace traditional DCF method practically due to its mathematical complexity. This paper introduces an option pricing valuation model specifically adjusted for the natural resource development projects. We add market information and industry-specific features into the model so that the model remains objective as well as realistic after the adjustment. The following two features of natural resource development projects take central parts in model construction; product price is a unique source of cash flow's uncertainty, and the projects have cost structure from capital-intense industry, in which initial capital cost takes most part of total cost during the projects. To improve the adaptability of Option Pricing Model specifically to the natural resource development projects, we use Two-Factor Model and Long-term Asset Model for the analysis. Although the model introduced in this paper is still simple and reflects limited reality, we expect an improvement in applicability of option pricing method for the evaluation of natural resource development projects can be made through the process taken in this paper.

• Journal of Korean Elementary Science Education
• /
• v.23 no.4
• /
• pp.344-356
• /
• 2004

In this study, we developed the program for 'Discretionary Activity' focusing on the multiple activities with everyday-life materials to enhance scientific creativity (MAEM-SC), which was specifically for students in the 6-7th grade according to the 7th curriculum in Korea. As important factors for scientific creativity, we selected the ability to find out the context relevant to scientific problems, the ability to connect the problem context to scientific knowledge, the ability to invent the ways to solve the problem scientifically, and ability to concentrate on the scientific problem solving activity. The topics of the program were drawn from common and familiar things in our everyday contexts, such as human body, everyday tools, food, play and toys, and everyday episodes. The multiple activities here mean the activities which are systematically constructed with the various types of activities with a specific intention. The multiple activities were designed in three types, that is, series type, parallel type, and combination type. Each of them consists of the several activities as follows: estimating and measuring, carrying out an experiment using body, inventing implement (tools), thinking statistically, writing creatively with scientific themes, and connecting one concept to another concept etc. Through a trial of the program, we found that this program has some positive influence on the enhancement both of the ability to find out the context relevant to scientific problems and the ability to connect it to the students' existing scientific knowledge.

• Information Systems Review
• /
• v.20 no.3
• /
• pp.51-71
• /
• 2018

Many studies in the field of information security reveal the need to increase awareness. However, although awareness of information security has been raised to a considerable extent, actual security behavior has been shown to fall short of that. Therefore, we wanted to identify the role of psychological factors in making information security decisions by conducting a experimental study. The results show that there are differences in perception of information security risks according to the probabilistic distance and the degree of relative optimism due to social distance. In relation to their relative optimism and intention of information security, they reduced the level of perceived risk compared to those close to them and found that their influence varied according to their probabilistic distance. This study has made valuable attempt in terms of methodology and it is meaningful that the psychological factor is taken into consideration for the information protection behavior, so that the range of relative optimism that actually affects the perception of risk is narrowed. It is expected to contribute to the improvement of information security level of information technology users and protection of information assets by empirically identifying necessity of various approaches to decision making process for information security.