• Title/Summary/Keyword: 베이지안 이론

Search Result 81, Processing Time 0.037 seconds

문항반응이론에서의 추정방법과 대입학력고사의 문항분석

  • 박정수;조완현
    • Communications for Statistical Applications and Methods
    • /
    • v.1 no.1
    • /
    • pp.192-205
    • /
    • 1994
  • 본 논문에서는 피험자의 능력과 검사문항에 정답할 확률과의 관계에 기초한 문항반응 이론의 기본 가정과 통계적 모형을 소개하였다. 또한 검사의 목적상 필요한 피험자의 능력을 정확히 추정하는 방법과, 검사에 사용되는 각 문항을 특성지우는 문항모수의 여러가지 통계적 추정 방법에 대하여 정리하였다. 그 방법들은 결합 최우추정법, 조건부 최우추정법, 주변 최우추정법, 베이지안 추정법 및 이들의 혼합에 의한 방법이다. 문항반응이론의 적용의 한 예로서 93년도 대입학력고사의 수학 시험문항을 BILOG 라는 컴퓨터 프로그램을 이용하여 분석하였다.

  • PDF

Bayesian Analysis for the Zero-inflated Regression Models (영과잉 회귀모형에 대한 베이지안 분석)

  • Jang, Hak-Jin;Kang, Yun-Hee;Lee, S.;Kim, Seong-W.
    • The Korean Journal of Applied Statistics
    • /
    • v.21 no.4
    • /
    • pp.603-613
    • /
    • 2008
  • We often encounter the situation that discrete count data have a large portion of zeros. In this case, it is not appropriate to analyze the data based on standard regression models such as the poisson or negative binomial regression models. In this article, we consider Bayesian analysis for two commonly used models. They are zero-inflated poisson and negative binomial regression models. We use the Bayes factor as a model selection tool and computation is proceeded via Markov chain Monte Carlo methods. Crash count data are analyzed to support theoretical results.

Bayesian Computation for Superposition of MUSA-OKUMOTO and ERLANG(2) processes (MUSA-OKUMOTO와 ERLANG(2)의 중첩과정에 대한 베이지안 계산 연구)

  • 최기헌;김희철
    • The Korean Journal of Applied Statistics
    • /
    • v.11 no.2
    • /
    • pp.377-387
    • /
    • 1998
  • A Markov Chain Monte Carlo method with data augmentation is developed to compute the features of the posterior distribution. For each observed failure epoch, we introduced latent variables that indicates with component of the Superposition model. This data augmentation approach facilitates specification of the transitional measure in the Markov Chain. Metropolis algorithms along with Gibbs steps are proposed to preform the Bayesian inference of such models. for model determination, we explored the Pre-quential conditional predictive Ordinate(PCPO) criterion that selects the best model with the largest posterior likelihood among models using all possible subsets of the component intensity functions. To relax the monotonic intensity function assumptions, we consider in this paper Superposition of Musa-Okumoto and Erlang(2) models. A numerical example with simulated dataset is given.

  • PDF

Variable Message Sign Operating Strategies Based on Bayesian Games (베이지안 게임이론에 근거한 전략적 VMS 제공에 관한 연구)

  • Kwon, Hyug;Lee, Seung-Jae;Shin, Sung-Whee
    • Journal of Korean Society of Transportation
    • /
    • v.22 no.7 s.78
    • /
    • pp.71-78
    • /
    • 2004
  • This paper presents a game-theoretic model of information transmission for variable message sign(VMS) operations. There are one VMS operator and many drivers as players. Operator wants to minimize the total travel time while the drivers want to minimize their own travel time. The operator who knows the actual traffic situation offers information strategically. The drivers evaluate the information from operator, and then choose the route. We model this situation as a cheap-talk game which is a simplest form of Bayesian game. We show that there is a possibility that the operator can improve the traffic efficiency by manipulating the electric signs at times. Indeed, it is an equilibrium of the game. This suggests that the operator must consider the strategic use of VMS system seriously.

Selective Inference in Modular Bayesian Networks for Lightweight Context Inference in Cell Phones (휴대폰에서의 경량 상황추론을 위한 모듈형 베이지안 네트워크의 선택적 추론)

  • Lee, Seung-Hyun;Lim, Sung-Soo;Cho, Sung-Bae
    • Journal of KIISE:Software and Applications
    • /
    • v.37 no.10
    • /
    • pp.736-744
    • /
    • 2010
  • Log data collected from mobile devices contain diverse and meaningful personal information. However, it is not easy to implement a context-aware mobile agent using this personal information due to the inherent limitation in mobile platform such as memory capacity, computation power and its difficulty of analysis of the data. We propose a method of selective inference for modular Bayesian Network for context-aware mobile agent with effectiveness and reliability. Each BN module performs inference only when it can change the result by comparing to the history module which contains evidences and posterior probability, and gets results effectively using a method of influence score of the modules. We adopt memory decay theory and virtual linking method for the evaluation of the reliability and conservation of casual relationship between BN modules, respectively. Finally, we confirm the usefulness of the proposed method by several experiments on mobile phones.

2005학년도 대학수학능력시험 '수리영역 가형'에 대한 문항분석

  • Lee, Gang-Seop;Kim, Jong-Gyu
    • Communications of Mathematical Education
    • /
    • v.19 no.1 s.21
    • /
    • pp.321-323
    • /
    • 2005
  • 본 연구에서는 2004년 11월에 시행된 '2005학년도 대학수학능력시험 수리영역 가형' 의 문항을 분석하였다. 즉, 2-모수 문항 반응 모형에 근거한 베이지안(Bayesian) 1.0을 이용하여 문항의 난이도 및 변별도를 측정하였으며 고전검사이론 프로그램임 테스트안(Testan) 1.0을 이용하여 문항의 신뢰도 및 오답지 매력도를 구하였다. 이 결과는, 학생들이 어느 단원을 어려워하고 어떤 내용을 이해하지 못 하는지 그 원인을 찾을 수 있으므로, 교수-학습의 기초 자료로 활용할 수 있을 것이다.

  • PDF

Theoretical Considerations for the Agresti-Coull Type Confidence Interval in Misclassified Binary Data (오분류된 이진자료에서 Agresti-Coull유형의 신뢰구간에 대한 이론적 고찰)

  • Lee, Seung-Chun
    • Communications for Statistical Applications and Methods
    • /
    • v.18 no.4
    • /
    • pp.445-455
    • /
    • 2011
  • Although misclassified binary data occur frequently in practice, the statistical methodology available for the data is rather limited. In particular, the interval estimation of population proportion has relied on the classical Wald method. Recently, Lee and Choi (2009) developed a new confidence interval by applying the Agresti-Coull's approach and showed the efficiency of their proposed confidence interval numerically, but a theoretical justification has not been explored yet. Therefore, a Bayesian model for the misclassified binary data is developed to consider the Agresti-Coull confidence interval from a theoretical point of view. It is shown that the Agresti-Coull confidence interval is essentially a Bayesian confidence interval.

Bookmark Classification Agent Based on Naive Bayesian Learning Method (나이브 베이지안 학습법에 기초한 북마크 분류 에이전트)

  • 최정민;김인철
    • Proceedings of the Korea Multimedia Society Conference
    • /
    • 2000.04a
    • /
    • pp.405-408
    • /
    • 2000
  • 최근 인터넷의 발전으로 많은 정보와 지식을 우리는 인터넷에서 제공받을 수 있게되었다. 인터넷에 존재하는 정보는 수많은 웹서버에 산재되어 있으며, 정보의 위치는 주소(URL)를 가지고 존재하게 되는데 사용자는 자신이 관심있는 정보의 주소를 저장하기 위하여 웹브라우저 북마크(Bookmark)기능을 사용한다. 그러나 북마크 기능은 웹문서의 주소 저장에 일차적인 목적을 두고 있으며, 이후 북마크의 개수가 증가하면, 사용자는 북마크관리가 어렵게되므로 사용자 북마크 파일을 자동으로 분류하여 관리할수 있는 에이전트 기술을 사용하고자 한다. 대표적인 분류에이전트 시스템으로는 전자우편 분류 에이전트인 Maxims, 뉴스기사 분류 에이전트인 NewT, 엔터테인먼트(Entertainment) 선별 에이전트인 Ringo 등이 있다. 이러한 시스템들은 분류할 대상에 따라 조금씩 다른 모습의 에이전트 기능을 보이고 있으며, 본 논문은 기계학습 이론중 교사학습 알고리즘인 나이브 베이지안 학습방법(Naive Bayesian Learning method)을 사용하여 사용자가 분류하지 못한 북마크를 자동으로 분류하는 단일 에이전트 기반 북마크 분류기를 설계, 구현하고자한다.

  • PDF

Bayesian Inference for Mixture Failure Model of Rayleigh and Erlang Pattern (RAYLEIGH와 ERLANG 추세를 가진 혼합 고장모형에 대한 베이지안 추론에 관한 연구)

  • 김희철;이승주
    • The Korean Journal of Applied Statistics
    • /
    • v.13 no.2
    • /
    • pp.505-514
    • /
    • 2000
  • A Markov Chain Monte Carlo method with data augmentation is developed to compute the features of the posterior distribution. For each observed failure epoch, we introduced mixture failure model of Rayleigh and Erlang(2) pattern. This data augmentation approach facilitates specification of the transitional measure in the Markov Chain. Gibbs steps are proposed to perform the Bayesian inference of such models. For model determination, we explored sum of relative error criterion that selects the best model. A numerical example with simulated data set is given.

  • PDF

Fingerprinting Bayesian Algorithm for Indoor Location Determination (실내 측위 결정을 위한 Fingerprinting Bayesian 알고리즘)

  • Lee, Jang-Jae;Kwon, Jang-Woo;Jung, Min-A;Lee, Seong-Ro
    • The Journal of Korean Institute of Communications and Information Sciences
    • /
    • v.35 no.6B
    • /
    • pp.888-894
    • /
    • 2010
  • For the indoor positioning, wireless fingerprinting is most favorable because fingerprinting is most accurate among the technique for wireless network based indoor positioning which does not require any special equipments dedicated for positioning. The deployment of a fingerprinting method consists of off-line phase and on-line phase and more efficient and accurate methods have been studied. This paper proposes a bayesian algorithm for wireless fingerprinting and indoor location determination using fuzzy clustering with bayesian learning as a statistical learning theory.