Semantic and Episodic Memory, Same or Different Category?: With Reference to the TV Documentary Ancient Aliens

노윤아 영상영어교육학회 2016 영상영어교육 (STEM journal) Vol.17 No.4

The purpose of the paper is to investigate how semantic information (=SI) and episodic information (=EI) interact from an educational perspective. Tulving (2002) argues that semantic memory (=SM) and episodic memory (=EM) are separate categories of long-term memory (=LTM). In this paper, we consider that Tulving’s argument may be inappropriate in educational contexts, that SI and EI exist on a continuum in LTM, and that SM is not a separate category. The subjects were four college students majoring in English. They used the TV documentary Ancient Aliens (Burns, 2009), which is full of SI. In the first stage of the experiment, all the subjects wrote short reports about content they found hard to understand. Next, they were required to select impressive expressions and unimpressive ones and to write comments on their selections. After one month, a delayed recognition test was administered. The comment writing activity helped convert SI into EI. The result was that they improved their degree of memory for both selections. This experiment shows that humans can understand SI by way of EI. This has implications for education in that teachers need to develop activities in which SI is converted into EI in order to help students understand clearly.

How Do We Define Authenticity? A Comparison of a Corpus with the TV Drama, Grimm

노윤아 영상영어교육학회 2012 영상영어교육 (STEM journal) Vol.13 No.2

The aim of this study is to examine the authenticity of a corpus by comparing it with an American TV drama. Through the results of the comparison, we can find out what is perceived as authentic by learners and what is more appropriate to suggest as a conversation model. The role of the content and form in making conversations authentic to learners is emphasized and these have been chosen as criteria for comparison. The corpus of contemporary American English and the American TV drama, Grimm were used as materials for the comparison. According to the results, the content of COCA is realistic, but boring and hard to understand while the dramatized content of Grimm is interesting to students. In addition, the transformed form ignoring the discourse units of COCA is not appropriate to suggest a conversation model compared to the preserved form following the discourse units of Grimm. These findings suggest that the realism of materials is not enough to obtain authenticity and that scripted and dramatized materials might be perceived as more authentic to learners.

혼돈/복잡성 이론(Chaos/Complexity Theory) 관점에 따른 영화 영어 학습에서의 ‘Interaction’ 연구

노윤아 영상영어교육학회 2019 영상영어교육 (STEM journal) Vol.20 No.1

This study had two aims. First, it sought to offer a new perspective on interaction by following Larsen-Freeman’s (1997) chaos/complexity theory (C/CT). Second, it set out to examine the types of interactions that occur when movies are used as English teaching materials. In this study, four college student participants received instruction based on the American movie I Feel Pretty (Kohn & Silverstein, 2018). To observe the participants’ interactions, the researchers asked them to complete the following five tasks: describe the movie’s plot, compare it with another similar Korean movie, discuss culture, write a short essay about the movie in English, and discuss multi-word units (MWUs). The first three tasks concerned content and the researchers expected all participants to actively interact with each other when completing them. All the participants actively discussed culture; however, they did not enjoy talking about the other two content-related topics. The final two tasks pertained to language. In completing these tasks, the participants barely interacted with each other and mainly interacted with the teacher. Also, the participants were very interested in learning MWUs. In sum, the participants’ interactions did not match the teacher’s expectations overall, indicating that Larsen-Freeman’s C/CT most persuasively explains interactions in real-life English class situations. .

Attention LSTM과 SHAP을 사용한 설명 가능한 COVID-19 확진자 수 예측 기법

노윤아,정승원,문재욱,황인준 한국정보과학회 2021 데이타베이스 연구 Vol.37 No.2

Due to the spread of COVID-19, many losses are being reported worldwide. In order to establish effective countermeasures to prevent the spread of COVID-19, it is necessary to accurately predict the extent of the spread of COVID-19. However, due to the recent introduction of vaccines, a forecasting model suitable for the changed diffusion pattern is needed, and although the forecasting accuracy has been improved through a machine learning-based approach, the model does not provide sufficient confidence due to the lack of explanatory properties of the model. Therefore, in this paper, we propose an explainable forecasting scheme that forecasts the number of COVID-19 confirmed cases using an Attention LSTM(Long Short-Term Memory) model and explains the analysis results using SHAP(SHapley Additive exPlanations). Changes due to introduction of vaccines were reflected by collecting various data related to COVID-19, including vaccination data, and using it as an input variable for the Attention LSTM model. Through comparative experiments with various models, we demonstrated the excellent forecasting performance of the proposed model and the explanatory capacity of the SHAP for the results. COVID-19의 확산으로 인해 세계적으로 많은 손실이 보고되고 있다. 효과적인 확산 방지 대책을 수립하기 위해서는 COVID-19 확산 정도의 정확한 예측이 필요하며, 이를 위해 확산 초기부터 기계적 접근법이나 기계학습 기반 접근법 등을 활용한 기법들이 제안되어왔다. 그러나 최근 백신 도입으로 인하여 변화된 확산 패턴에 적합한 예측 모델이 필요하며, 기계학습 기반 접근법을 통해 예측 정확도는 높아졌으나 모델의 설명성이 부족하여 충분한 신뢰를 주지 못하는 실정이다. 이에, 본 논문에서는 Attention LSTM(Long Short-Term Memory) 모델을 사용하여 COVID-19 확진자 수를 예측하고, 그 결과를 SHAP(SHapley Additive exPlanations)을 통하여 분석하는 설명 가능한 COVID-19 확진자 수 예측 기법을 제안한다. 백신 접종 데이터를 포함한 COVID-19 관련 다양한 데이터를 수집하고, Attention LSTM 모델의 입력 변수로 사용함으로써 백신 도입으로 인한 변화를 반영하였다. 다양한 모델과의 비교 실험을 통하여 제안한 모델의 우수한 예측 성능을 보였으며, SHAP을 통해 예측 결과에 대한 설명 가능성을 입증하였다.

왜 미국 드라마를 사용하는가?: 대화자각력을 중심으로

노윤아 영상영어교육학회 2010 영상영어교육 (STEM journal) Vol.11 No.2

This study used American TV dramas to raise students' conversational awareness. Conversational competence should be distinguished from language competence. In most English conversation classes, however, the two competencies are often confused. A lot of activities which are designed to strengthen conversational abilities are in fact only language practice. Students are unaware of conversational structures and strategies and therefore don't fully understand the nature of conversation. Participants in the study were 40 college students. The students analyzed the structures and strategies of conversations which appeared in scripts for TV dramas. To collect data, participants completed a questionnaire and were tested on conversational analysis. The data showed that students who didn't express any interest in conversation began to give attention to conversational structures and strategies. They were also generally positive toward the way of learning conversation through the use of American TV dramas. Conversational awareness can be developed by improving students' understanding of the nature of conversation, its structures, and its strategies. Using American TV dramas could be an excellent way to strengthen conversational competence.

영화를 활용한 읽기활동 연구

노윤아 영상영어교육학회 2008 영상영어교육 (STEM journal) Vol.9 No.2

This article suggests a variety of suitable reading activities that can be used as learning materials related to movies in English. Reading is an essential activity for language input but today's learners often tend to disregard reading and emphasize speaking and listening. Even worse, they try to avoid reading altogether. Teachers and researchers therefore tried to find a way to motivate and interest learners to use suitable reading activities related to movies in English. There are few studies which suggest specific reading activities related to movies so the teachers who want to use movies in their English reading classes have trouble finding practical guides for teaching. By showing some possible reading activities, this study will be a help to them.

EFL 학습자들의 유창한 발화를 위한 절차기억 활성화 방안 연구: 암기와 반복을 중심으로

노윤아 한국현대영어영문학회 2019 현대영어영문학 Vol.63 No.4

This study’s aim is to identify and suggest a method for improving EFL learners’ fluency by activating their procedural memory through memorization and repetition activities. Improving L2 English learners’ fluency requires the frequent use in EFL classes of activities that depend on the procedural memory system, which underlies the implicit learning and processing of a wide range of perceptual-motor and cognitive skills (Ullman & Lovelett, 2016, p. 3). Four college students, all English majors with advanced reading skills, participated in this research. They selected three scenes from the movie 500 Days of Summer (Webb, 2009) and discussed them through the lens of their personal experiences with relationships. Subsequently, participants completed three rounds of memorization and repetition activities with one-week intervals between each round. The participants were required to express the characters’ feelings by memorizing and repeating the selected scenes. The study’s results indicate that all participants were initially reluctant to attempt memorization and repetition activities; however, after trying them, they became more willing to do so. Moreover, their attitudes towards memorizing and repeating became more positive. The participants’ mistakes in memorizing the movie scenes were reduced throughout the research process. Therefore, introducing more memorization and repetition activities, which have the capacity to activate students’ procedural memory, will improve the fluency of L2 learners.

대학생을 위한 노래를 활용한 다단어 단위(multi-word units) 학습법 제안

노윤아 한국외국어교육학회 2018 한국외국어교육학회 학술대회 자료집 Vol.2018 No.-

햐쿠타 나오키(百田尙樹)의 『영원한 제로(永遠の0)』와 『해적이라 불린 사나이(海賊とよばれた男)』고찰

노윤선(盧潤宣) 한국일본문화학회 2018 日本文化學報 Vol.0 No.78

Since the 2000s, a number of Japanese works in different genres, such as literature, films, and animations, have presented right-leaning views that mirror the nation’s conservative swing of politics that started in the 1990s. This paper considers the historical and ethical perspectives of two literary works by Naoki Hyakuta, who gained huge popularity by creating several conservative-oriented works as part of this cultural and artistic movement. In 2006, Hyakuta published The Eternal Zero, the story of Kamikaze pilots in the Second World War; he became a best-selling author when it sold about 5.5 million copies in Japan alone. After that, he created many other works. This paper reviews Hyakuta’s literary works of the 2000s, by understanding the historical and ethical perspectives in his works, which tend to rationalize the Japanese wars and colonial rule of the past.

메타 학습을 활용한 신종 감염병 예측 기법

노윤아,정승원,박성우,황인준 한국정보과학회 2022 정보과학회 컴퓨팅의 실제 논문지 Vol.28 No.11

To reduce damage caused by infectious diseases, it is important to forecast the extent of the spread of infectious diseases and prepare countermeasures accordingly. So far, various forecasting models have been proposed to forecast the number of future confirmed cases. However, in the case of emerging infectious diseases, as it is difficult to collect a sufficient amount of occurrence data for forecasting, it is a challenging task to construct an effective forecasting model. In this paper, we proposed a meta-learning-based new infectious disease forecasting scheme that could achieve a high forecasting accuracy even when the number of data was insufficient. To this end, we first collected a large volume of time-series data from diverse domains. We then, extracted various meta-data from each time-series data and determined the optimal forecasting model and hyperparameters using these data. Finally, we carried out meta-learning to train our forecasting model. To demonstrate the excellence of our proposed forecasting scheme, we performed various experiments using COVID-19 data assuming various initial states of disease development according to the period. 감염병의 피해를 줄이기 위해서는 감염병 확산 정도를 예측하고 그에 따른 대응책을 마련하는 것이 중요하다. 이제까지 미래 감염자 수를 예측하기 위한 다양한 예측 모델이 제안되었다. 그러나 신종 감염병의 경우, 예측에 필요한 충분한 양의 발생 데이터 수집이 어려워 효과적인 예측 모델 구성이 쉽지 않다. 본 논문에서는 데이터의 수가 부족한 경우에도 높은 예측 정확도를 달성할 수 있는 메타 학습 기반의 신종 감염병 예측 기법을 제안한다. 이를 위해 다양한 시계열 데이터로부터 각 시계열 데이터의 특징 정보와 최적의 예측 모델 및 하이퍼-파라미터 정보를 담고 있는 메타데이터를 확보하고 이들을 기반으로 모델을 훈련하는 메타 학습을 진행한다. 적은 양의 신종 감염병 데이터를 대상으로 특징 정보를 추출하고 학습된 모델의 입력으로 사용하여 최적의 예측 모델을 도출한다. 최근 수집된 코로나19 데이터를 이용하여, 기간에 따른 다양한 발생 초기 상태를 가정한 실험을 통해 제안한 기법의 우수성을 입증한다.