http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Haiku, Zen, and Wanaki: The Hybridized Poetics of Gerald Vizenor
( Kim¸ Seonghoon ) 동국대학교 영어권문화연구소 2017 영어권문화연구 Vol.10 No.3
This paper explores how Anishinaabe writer and postmodern critic Gerald Vizenor engages with creating animated words that concern trans-cultural conversations among different literary traditions. Both of Vizenor’s fictional and poetic works show an attempt to negotiate between Japanese culture and indigenous oral tradition and even between Western thoughts and indigenous philosophy. This is possible because Vizenor’s works represent the process in which his animated words, like native trickster figures, move beyond any fixed identity or time and space, actively participating in language games. While focusing on Vizenor’s theoretical commitment, this paper deals with Vizenor’s novel Dead Voices (1992), where Vizenor’s fictional device, Wanaki game, represents how trickster figures from Anishinaabe oral tradition work, and analyzes some of his early poems which concern the themes of Haiku based on Zen philosophy and Ojibwe dream songs. In demonstrating how such a hybridized voice is heard in Vizenorian poetry, “The Balcony,” one of Vizenor’s first serious poems is particularly compared to “On the road,” one of poems by Matsuo Basho, the Haiku master.
Seonghoon Jin,Hong Shick Min,Young June Park 한국물리학회 2004 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.44 No.1
We propose a new transport model, named the `hydrodynamic density-gradient (HDG) model', that can simulate the quantum-mechanical and the nonlocal transport effects simultaneously. The HDG model is a hydrodynamic expansion of the density-gradient model that treats the quantum effects by using a quantum correction term. The governing equation set of the HDG model includes the Poisson equation, the electron continuity equation, the quantum potential equation, and the energy balance equation. To discretize the governing equations, we apply the control volume method with nonlinear discretization schemes for the electron and the hole ux terms. To implement the HDG model, we have developed a generalized multi-dimensional device simulator called NANOCAD. We also have implemented other transport models such as the coupled Schrodinger-Poisson model, the drift-diffusion model, the density-gradient model, and the hydrodynamic model to compare them with the HDG model. We have simulated a 25-nm NMOSFET device, and the results show that our model successfully predicts the quantum-confinement effects, the lateral quantum effects, and the velocity overshoot effects in the device.
A Study on IRT Fixed Parameter Calibration Methods Using BILOG-MG
Seonghoon Kim(김성훈) 한국교육평가학회 2006 교육평가연구 Vol.19 No.1
공통 능력 척도의 개발은 문항반응이론(IRT)의 모수불변성을 잘 활용하기 위해서 여러 응용 연구에서 요구된다. 그 개발을 위해 세 가지 추정 방법이 사용될 수 있다: 다집단 동시추정 방법, 개별 추정 후 연계(linking) 방법, 고정모수 추정 방법. 본 연구의 목적은 IRT 추정 프로그램인 BILOG-MG에서 기본적으로 제공 되는 고정모수 추정 기능을 진단하고, 그 기능을 올바로 사용하기 위한 두 가지 대안적인 추정 방식을 제안하는 데 있다. 이를 위해 세 가지 고정모수 추정 방법이 고려되었다: 사전능력분포 비갱신 방법, 사전능력분포 단순 갱신 방법, 사전능력분포 반복갱신 방법. 비교의 목적상, 동시 추정 방법과 개별 추정 후 연계 방법 또한 고려되었다. 사전분포 비갱신 방법은 BILOG-MG에서 기본적으로 제공되는 방법이다. 사전능력분포 단순갱신 방법은 해당 분석 자료에 예비적으로 개별 추정 방법을 적용하여 나오는 사후능력분포를 분석 집단의 능력 척도에 맞게 선형 변환한 후 그것을 본 고정모수 추정 방법의 사전능력분포로 사용해서 비고정 (즉, 비공통) 문항의 모수를 추정하는 방법이다. 사전능력분포 반복갱신 방법은 BILOG-MG를 고정모수 추정 방식으로 반복적으로 돌려서 갱신된 사전능력분포를 사용하여 비고정 문항의 모수를 추정하는 방법이다. 세 가지 다른 정규분포-N(-1, 1), N(0, 1), N(1, 1)-각각으로부터 나온 컴퓨터 모의 피험자들을 사용해서 산출된 문항반응 자료를 통해 세 가지 고정모수 추정 방법을 진단하였다. 연구 결과, N(0, 1) 분포를 통해 나온 검사자료에 대해서는 세 가지 방법 모두 잘 작용하는 것이 확인되었다. 사전능력분포 단순갱신 방법과 반복갱신 방법은 검사자료가 N(-1, 1)과 N(1, 1) 분포를 통해 나온 경우에도 잘 작용하는 것으로 나타났다. 이와 대조적으로, 비갱신 방법은 N(-1, 1) 분포로부터 나온 자료에 대해서는 문항곤란도 모수를 과대 추정하는 경향을, 그리고 N(1, 1) 분포로부터 나온 자료에 대해서는 그 모수를 과소 추정하는 경향을 보였다. Many applications of item response theory (IRT) require development of a common ability scale. Three calibration methods can be used for this purpose: (1) separate calibration with linking, (2) concurrent calibration, and (3) fixed parameter calibration (FPC). This paper diagnoses the FPC facility of the IRT estimation program BILOG MG 7.0 and presents two alternative FPC methods that can be suitably used under the current limitation for FPC using the program. Three FPC methods were considered: (1) no prior update (NPU), (2) simple prior update (SPU), and (3) iterative prior update (IPU) methods. For comparative purposes, concurrent and separate calibrations were considered as well. The NPU method is the default FPC method supported by BILOG MG 7.0. The SPU method uses as a prior a linear transformation of the posterior ability distribution from a preliminary separate calibration with the data used to conduct FPC (so that it fits the FPC examinees). The IPU method uses iteratively updated prior ability distributions through multiple runs to conduct FPC. Performances of the three FPC methods were examined with test data generated using simulees from three normal distributions: N(-1, 1), N(0, 1) and N(1, 1). All three FPC methods appeared to perform properly under the N(0, 1) distribution condition. The SPU and IPU methods also appeared to perform properly with data from the N(-1, 1) and N(1, 1) distributions. The NPU method, however, tended to overestimate difficulty parameters when response data were generated from the N(-1, 1) distribution, and to underestimate difficulty parameters when the N(1, 1) distribution was used for generating data.
( Seonghoon Kim ) 경남대학교 인문과학연구소 2018 人文論叢 Vol.45 No.-
This paper aims at demonstrating that the portraits of contemporary American Indians in his works resist the fossilized and stereotypical images of Indians, and even go beyond the ethnical boundaries set by ‘traditional’ American Indian literature. After sketching out the lines of criticism on seemingly stereotypical characters by some American Indian writers and scholars, this paper examines how the Spokane writer engages with the controversial idea of ‘Indian’ and ‘Indianness’ in his fiction, especially in The Toughest Indian in the World (2000). As this paper proposes, Alexie’s unflinchingly bold representation of new, unexpected Indian characters informs his serious quest for Indian identity. It should be also considered as a literary strategy because they illustrate ways of resisting any easy definitions or colonial labels of contemporary Native people in the US. For example, Etta Joseph in “Dear John Wayne” is subversive in that her story twists a colonial relationship between a white man and an ‘Indian princess.’
Increased Eciency of Range Verification in Routine QA for Pencil-Beam Scanning Proton Therapy
Seonghoon Jeong,Myonggeun Yoon,Jaeman Son,Kwangzoo Chung 한국물리학회 2018 THE JOURNAL OF THE KOREAN PHYSICAL SOCIETY Vol.73 No.7
Quality assurance (QA) is required when performing pencil-beam scanning proton therapy, but the eciency of QA is degraded in proportion to the energy of the protons. We developed a method to assess the preferred energy range and distal fall-off by combining multiple Bragg peaks to increase the eciency of QA. Beams of 70, 110, 150, 190, and 230 MeV for exposure were planned using a treatment planning system. The Bragg curves for therapeutic proton beams were modeled using three different fitting function models, allowing the feasibility of a simple modeling of the Bragg curve to be investigated. The planned beams were exposed and measured using a multi-layered ionization chamber. Software developed using a Python tool could detect five Bragg peaks from the integrated curves that were fitted based on polynomial, cubic spline and Landau distributions. This software could calculate the range and distal fall-off of the five fitted peaks. For the verification of the accuracy of this method, the calculated results were compared with the range and distal fall- off obtained by exposing and analyzing five single-energy beams individually. Comparisons of the Bragg peaks for the five energies exposed individually with the results obtained by exposing them all at once showed that the ranges of the energy beams when using the polynomial fitting and the cubic spline modes were 0.16 mm and 0.10 mm longer, respectively, while the distal fall-offs were 0.14 mm and 0.06 mm shorter, respectively. When using the Landau distribution fitting, the range was 0.06 mm longer and the distal fall-off was 0.04 mm shorter. Analyses of the ranges and distal fall-offs of the five energy beams exposed at once with single-beam loading by using the method developed in this study showed no significant differences from the results obtained by exposing the energy beams individually. Thus, range verification QA by using the proposed method is not only suitable for single-proton beams with multiple energies but also reduces the measurement time.