• Journal of KIISE:Software and Applications
• /
• v.34 no.11
• /
• pp.1001-1010
• /
• 2007

HMM tends to output high probability for not only the proper class data but confusable class data, since the modeling power increases as the number of parameters increases. Thus it may not be helpful for discrimination to simply increase the number of parameters of HMM. We proposed two methods in this paper. One is a CMC(Confusion Likelihood Model Selection Criterion) using confusion class data probability, the other is a new recognition method, RCM(Recognition Using Confusion Models). In the proposed recognition method, confusion models are constructed using confusable class data, then confusion models are used to depress misrecognition by confusion likelihood is subtracted from the corresponding standard model probability. We found that CMC showed better results using fewer number of parameters compared with ML, ALC2, and BIC. RCM recorded 93.08% recognition rate, which is 1.5% higher result by reducing 17.4% of errors than using standard model only.

• Journal of KIISE:Software and Applications
• /
• v.30 no.9
• /
• pp.867-875
• /
• 2003

This paper concerns continuous density HMM topology optimization. There have been several researches for HMM topology optimization. BIC (Bayesian Information Criterion) is one of the well known optimization criteria, which assumes statistically well behaved homogeneous model parameters. HMMs, however, are composed of several different kind of parameters to accommodate complex topology, thus BIC's assumption does not hold true for HMMs. Even though BIC reduced the total number of parameters of HMMs, it could not improve the recognition rates. In this paper, we proposed two new model selection criteria, HBIC (HMM-oriented BIC) and BIC_Anti. The former is proposed to improve BIC by estimating model priors separately. The latter is to combine BIC and anti-likelihood to accelerate discrimination power of HMMs. We performed some comparative research on couple of model selection criteria for online handwriting data recognition. We got better recognition results with fewer number of parameters.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.10d
• /
• pp.571-573
• /
• 2002

음성인식과 온라인 필기인식에서 우수한 성능을 보이는 은닉 마르코프(HMM)의 HMM의 구조는 휴리스틱 한 방법에 의해 결정되는 것이 일반적이기 때문에 최적의 모델을 선택하는데 어려움이 있다. 이에 본 논문에서는 HMM의 구조를 체계적인 방법으로 정함과 동시에 변별력의 단점을 개선 할 수 있는 방법으로 Anti-likelihood를 이용한 모델간의 변별력을 살펴보고 최적의 모델 선택 기준인 BIC와의 결합하여, 체계적이고 효율적인 최적 모델 선택이 가능한 방법론에 대해 연구하고 필기데이터에 대해 검증한 결과, 기존의 방법보다 파라미터의 수는 감소되고 인식률이 향상됨을 알 수 있다.

• Journal of Soil and Groundwater Environment
• /
• v.12 no.1
• /
• pp.44-52
• /
• 2007

This study reviewed the overall process of application of contaminant fate and transport model as part of risk assessment. Site characterization and establishment of a conceptual model prior to establishing or selecting a appropriate model were described. Types of models, model selection guidance, and generic site conditions for model application were presented, the process of model calibration, validation, and sensitivity analysis were reviewed. Objectives of modeling should be defined before model selection, and the complexity of selected models should balance the quantity and quality of available input data with the desired model output. If model output is highly sensitive to an assumed or default value of input parameter, or fate and transport models cannot be adequately calibrated or validated, consideration should be given to other options such as using measured data or using another model.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.853-855
• /
• 2005

은닉 마르코프(HMM)의 HMM의 구조 최적을 위한 모델 선택 방법에 많은 방법들이 연구되어지고 있다. HMM의 구조를 어떻게 최적으로 정해야 하는 가에 대해 HMM의 구조를 체계적인 방법으로 정함과 동시에 변별력의 단점을 개선 할 수 있는 방법으로 Anti-likelihood(ALC1)를 제안하였고 이를 모델 선택 기준인 BIC와의 결합(ALC2)하여 필기 데이터에 대해 실험한 결과 기존의 방법보다 파라미터의 수는 감소되고 인식률이 향상됨을 알 수 있었다. 이를 Discrete HMM에도 적용하여 제안된 ALC2가 HMM 구조를 최적화하는 모델 선택 기준임을 Continuous-Density HMM과 비교하여 실험 검증 한다.

• Proceedings of the Korea Society of Mathematical Education Conference
• /
• 2010.04a
• /
• pp.217-217
• /
• 2010

다양한 선택이 가능한 대학 입시 선택 과정에서, 대학 및 학과별 선발 기준을 고려하여 최선의 선택을 해야 할 필요가 있다. 본 연구는 2010학년도 대학별, 학과별, 시기별 전형 기준을 근거로 이 과정을 수학적으로 모델링 한다. 그 예로 2010학년도 전형 기준을 수능 반영영역과 반영비율, 가중치 등을 고려하여 대학입시과정에서 활용될 수 있는 AHP를 통한 수학적 모델을 소개한다. 이 과정을 엑셀을 이용한 수학적 모델링으로 구현한다.

• Journal of The Korean Association For Science Education
• /
• v.34 no.2
• /
• pp.135-146
• /
• 2014

The purpose of this study is to explore evaluation criteria that pre-service elementary teachers employ as they evaluate and select models to explain electric circuits. Thirty junior students in a university of education have participated in the study as a part of the science education course in which they were enrolled. The lessons for the participants have been organized as a cyclic sequence of different modeling pedagogies including the expressive, experimental, and evaluative modeling. The pre-service teachers have been given five electric circuits in order and asked to create models and further develop them through peer discussion. Their modeling activities have been video- or audio-recorded, and the recordings and their transcripts have been analyzed using a framework of model evaluation criteria. It reveals that the types and frequencies of evaluation criteria used are different between situations of model development and model selection. While empirical and theoretical criteria have been used dominantly in both situations, more various criteria have been employed in the situation where the pre-service teachers selected one model among alternatives. Implications for science education and science education research have been suggested.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.29 no.2
• /
• pp.37-42
• /
• 2006

기술집약적인 경제성장의 중요성이 강조되고 있는 가운데 이공계 진학과 과학기술직종선택의 감소는 성장잠재력에 가장 근본적인 위협이 된다. 이를 유지하려는 여러 가지 정책이 교육학적이나 사회학적 근거에서 제시되고 있으나 이를 분석하는 이론적 경험적 틀이 상대적으로 부족한 상태이다. 직업선택모델은 사회학적인 접근법이 활발하게 진행되었으나, 경제적 동인에 대극 분석이 부족하다. 본 논문에서는 2000년 미국 센서스 데이터에 나타난 가장 국제화된 미국의 과학기술직 사례를 통해 인적자본 모델을 기준으로 하여 기술직에 대한 기존의 연구와 비교한다. 이민의 직업선택모델의 관점에서 원주민의 경우와 비교하며, 동시에 타 직업군과 비교한다. 직업선택에서 미래소득에 대한 예측이 대체로 정확하나 실제의 선택이 다르다는 기존 논문들의 주장은 성간 차이의 문제를 제외하고는 현격하지 않다. 민족적 차이의 효과도 인적자본효과에 비해서는 크지 않다. 과학기술직은 고급 화이트칼라 직종에 비해 결혼과 교육 언어 경험면에서 저급한 직종의 특성을 보인다. 여성의 과학기술직 기피는 남성프리미엄이 높아서는 아니지만 합리적인 차별 때문으로 볼 수 있다.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.14 no.1
• /
• pp.33-40
• /
• 2003

Radio frequency management system(RFMS) is being operated to facilitate national spectrum management and monitoring in Korea. To improve the engineering analysis function in RFMS, criteria for the automated selection of the propagation model adequate to the radio station service environment considered are proposed. Those criteria are derived from the specified parameters obtained through the analysis of related ITU-R propagation & diffraction loss models which are to be used in RFMS. Then, using criteria acquired, computer program is made to achieve the automated selection of the most appropriate propagation algorithm, among the ones provided in RFMS, to the environment in which the engineering analysis is required. Furthermore, an illustrative example is shown with the proposals fur increasing the efficiency of the engineering analysis in RFMS.

• Journal of KIISE
• /
• v.42 no.12
• /
• pp.1551-1560
• /
• 2015

Software Reliability Growth Models (SRGMs) are used to quantitatively evaluate software reliability and to determine the software release date or additional testing efforts using software failure data. Because a single SRGM is not universally applicable to all kinds of software, the selection of an optimal SRGM suitable to a specific case has been an important issue. The existing methods for SRGM selection assess the goodness-of-fit of the SRGM in terms of the collected failure data but do not consider the accuracy of future failure predictions. In this paper, we propose a method for selecting SRGMs using the trend of failure data and failure prediction ability. To justify our approach, we identify problems associated with the existing SRGM selection methods through experiments and show that our method for selecting SRGMs is superior to the existing methods with respect to the accuracy of future failure prediction.