http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Counting Process of MAP(3)s and Moment Fittings
김선교,Kim, Sunkyo The Korean Operations Research and Management Scie 2017 한국경영과학회지 Vol.3 No.1
Moments of stationary intervals and those of the counting process can be used for moment fittings of the point processes. As for the Markovian arrival processes, the moments of stationary intervals are given as a polynomial function of parameters whereas the moments of the counting process involve exponential terms. Therefore, moment fittings are more complicated with the counting process than with stationary intervals. However, in queueing network analysis, cross-correlation between point processes can be modeled more conveniently with counting processes than with stationary intervals. A Laplace-Stieltjies transform of the stationary intervals of MAP (3)s is recently proposed in minimal number of parameters. We extend the results and present the Laplace transform of the counting process of MAP (3)s. We also show how moments of the counting process such as index of dispersions for counts, IDC, and limiting IDC can be used for moment fittings. Examples of exact MAP (3) moment fittings are also presented on the basis of moments of stationary intervals and those of the counting process.
Generation of the Markovian Arrival Process of Order 2 Based on Joint Distribution
Sunkyo Kim(김선교) 대한산업공학회 2024 대한산업공학회지 Vol.50 No.2
In queueing network analysis, the Markovian arrival process, MAP, can be used as approximating arrival or departure processes. While the renewal processes such as Poisson process can be generated based on the marginal moments, the Markovian arrival process requires lag-1 joint moment or joint distribution function. In fact, the Markovian arrival process can be simulated by two transition rate matrices, (D0, D1), which is called the Markovian representation. However, finding a Markovian representation of higher order involves a numerically iterative procedure due to redundancy in the Markovian representation. Since there is one-to-one correspondence between joint moments and the coefficients of the joint Laplace transform, generating a MAP based on joint distribution can be much less complicated. In this paper, we propose an approach to generate a MAP based on joint distribution function which can be quickly obtained from joint moments and joint Laplace transform. Closed form formula and streamlined procedures are given for the simulation of MAP of order 2.
Sunkyo Kim(김선교) 한국경영과학회 2021 韓國經營科學會誌 Vol.46 No.3
Markovian arrival processes (MAPs) can be represented in many different ways such as the Markovian representation, Laplace transform, Jordan representation, and minimal realization problem (MRP) representation, to name a few. The MRP representation is given in two real-valued matrices (K’, R’) and can be used to determine the Markovian representation (D0, D₁) by similarity transformation. The MRP representation has been claimed to be unique but redundant. In this paper, we show that the MRP representation is not unique and then provide a non-redundant MRP representation of MAP(2)s. We also present closed-form formulas for the transformation from the MRP representation to the Markovian representation (D0, D₁) for MAP(2)s.
김선교(Kim Sunkyo),임경태(Lim KyungTae) 한국통신학회 2022 한국통신학회 학술대회논문집 Vol.2022 No.2
본 논문은 기관평가 시 평가자의 평가의견을 자동으로 요약하여 평가에 따른 보완사항을 생성하는 자연어처리 시스템을 제안한다. 기관 평가는 여러 정량/정성적 절차를 거쳐 이루어지며 그 중 중간평가로 평가의원의 의견이 도출된다. 본 논문에서는 해당 평가의견을 입력으로 받아 보완사항을 생성하기 위해 텍스트 생성모델에 활용되는 sequence-to-sequence 모델의 활용 방법을 제안한다.
Counting Process of MAP(3)s and Moment Fittings
Sunkyo Kim(김선교) 한국경영과학회 2017 한국경영과학회지 Vol.42 No.1
Moments of stationary intervals and those of the counting process can be used for moment fittings of the point processes. As for the Markovian arrival processes, the moments of stationary intervals are given as a polynomial function of parameters whereas the moments of the counting process involve exponential terms. Therefore, moment fittings are more complicated with the counting process than with stationary intervals. However, in queueing network analysis, cross-correlation between point processes can be modeled more conveniently with counting processes than with stationary intervals. A Laplace-Stieltjies transform of the stationary intervals of MAP (3)s is recently proposed in minimal number of parameters. We extend the results and present the Laplace transform of the counting process of MAP (3)s. We also show how moments of the counting process such as index of dispersions for counts, IDC, and limiting IDC can be used for moment fittings. Examples of exact MAP (3) moment fittings are also presented on the basis of moments of stationary intervals and those of the counting process.
김선교 ( Sunkyo Kim ),박준호 ( Joonho Park ) 한국정보처리학회 2020 한국정보처리학회 학술대회논문집 Vol.27 No.2
음성인식은 음향모델, 언어모델, 디코더 등의 기술을 이용한다. 음성인식은 하드웨어와 소프트웨어 구성이 정확하게 설계가 되어야 한다. 음성인식 프로젝트는 인프라 구성과 도입되는 음성인식 엔진도입, 인식률 그리고 시스템과의 연계가 중요하다. 하지만 음성인식 프로젝트는 솔루션 도입으로 인지하고 수행할 경우에는 많은 위험이 발생한다. 이 중 가장 문제가 되는 것이 인식률이다. 본 논문에서 음성 인식 개발 프로젝트 수행에 도출되는 인식률을 개선하는 방안을 제시하겠다.