http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
양구 DMZ 생태 관광자의 환경 인식이 친환경 관광 태도, 장소 애착, 정서적 연대 및 재방문 의도에 미치는 영향
노준호,이계희,정다은,노한나 한국관광연구학회 2023 관광연구저널 Vol.37 No.5
Since the Ministry of Environment and the Ministry of Culture, Sports, and Tourism jointly launched their plan to resuscitate ecotourism in 2008, domestic ecotourism has risen significantly in popularity. To prevent difficulties that may occur from insufficient environmental awareness or carelessness on the part of tourists visiting ecologically sensitive DMZ areas, it is important to analyze visitors' environmental consciousness. Therefore, the purpose of this study was to examine the relationship between ecotourists' environmental perceptions and eco-friendly tourism attitudes in terms of place attachment and emotional solidarity among tourists and their intention to revisit a DMZ ecotourism destination in Yanggu. People who had visited the DMZ ecotourism site in Yanggu comprised the on-site and online survey sample for this study. The empirical research revealed that environmental perception significantly influenced eco-friendly tourism attitudes, eco-friendly tourism attitudes significantly influenced place attachment, and place attachment significantly influenced emotional solidarity and intention to revisit. Finally, the emotional solidarity of tourists had a positive (+) effect on revisit intention. This study may contribute to the research on DMZ ecotourism destinations and may have implications for practitioners. In addition to academic contributions, practical significance is also highlighted.
노준호,김한준,장재영 한국전자거래학회 2012 한국전자거래학회 학술대회 발표집 Vol.2012 No.4
자동문서분류시스템의 성능을 높이기 위해서는 최적의 특징 집합을 구성하여 분류모델을 구축하는 것이 중요하다. 본 논문에서는 분류모델의 개선을 위해 워드넷(WordNet)을 이용하여 의미정보가 풍부한 특징을 생성하는 기법을 제안한다. 기본 아이디어는 학습문서집합 내에서 중요도가 높은 특징을 선정하여, 선정된 특징의 동의어 및 상위어를 추가함으로써 특징의 의미 확장을 수행하는 것이다. 또한 특징집합의 확장 가공을 위해 단어와 클래스간의 유사도를 계산함으로써 문서분류에 도움이 될 수 있는 단어들을 선별하였다. 결과적으로 제안한 특징 가중치 기법을 이용하여 나이브 베이즈 분류기의 성능을 개선하였다. 제안 기법을 평가하기 위해 표준 테스트 집합인 Reuters-21578 문서집합을 이용하여 실험을 수행하였다.
노준호,백혜정,Joon Ho Roh,Hae Jung Paik 대한안과학회 2008 대한안과학회지 Vol.49 No.7
Purpose: The purpose of this study is to analyze factors associated with recurrence and reoperation in intermittent exotropia. Methods: The medical records of 285 patients who had undergone the bilateral lateral recti recessions for intermittent exotropia with at least 1 year of postoperative follow-up were reviewed retrospectively. Surgical success was defined as a final deviation less than 10PD, overcorrection more than 10PD and recurrence less than 10PD at postoperative 3 months but recurring later, so being exodeviated more than 10PD in primary gaze at postoperative 1 year. Reoperation was performed when the eye was excessively exodeviated with follow-up, requiring surgical revision clinically. We investigated factors associated with recurrence, comparing recurrence group to success group, and also investigated factors associated with reoperation, comparing reoperation group to recurrence group. Results: Alignment at postoperative 1 day was proved to be significant factor influencing on recurrence. The age at the time of operation, preoperative alignment, stereopsis, etc were not associated with recurrence. Also, stereopsis was associated with reoperation, but there were no relationship between reoperation and other factors. Conclusions: Alignment at postoperative 1 day was related to recurrence and poor stereopsis (≥400 sec) was associated with reoperation. Therefore, sensory function, such as stereopsis should be considered more important than motor control when considering reoperation. J Korean Ophthalmol Soc 49(7):1114-1119, 2008
노준호(Jun-ho Roh),우승범(Seung-beom Woo),황원준(Won-jun Hwang) 한국정보통신학회 2021 한국정보통신학회 종합학술대회 논문집 Vol.25 No.1
지도학습에서 모델을 학습함에 있어 입력 데이터와 해당 데이터의 라벨이 필요하다. 하지만 신뢰성 있는 라벨링은 비용과 시간적인 면에서 많이 소요되며 이를 자동화할 경우 라벨이 언제나 맞는다는 보장이 없어 노이즈가 들어가게 된다. 이러한 라벨 노이즈 환경에서 지도학습을 진행할 경우 모델은 학습 초기에는 정확도가 올라가지만, 어느 정도 학습 후 정확도가 크게 감소되는 경향을 보인다. 라벨 노이즈 문제를 해결하기 위해 다양한 방법이 있지만, 대다수의 경우 모델이 예측한 확률을 수도라벨로 사용해 이용하는 경우가 많다. 여기에 대해서 우리는 모델이 예측한 확률을 정제하여 좀 더 빠르게 참라벨을 예측하는 방법을 제시한다. 기존의 논문 중 모델이 예측한 확률을 사용하는 방법에 우리가 제안하는 방법을 적용하여 같은 환경, 데이터셋에 대해 실험을 진행한 결과 성능개선과 더 빠르게 수렴하는 것을 확인할 수 있었다. 이를 통해 기존 연구들 중 모델이 예측하는 확률분포를 사용하는 방법들에 적용할 수 있고 같은 환경에서도 더 빠르게 수렴시킬 수 있기에 학습 소요시간을 줄일 수 있다. When learning a model in supervised learning, input data and the label of the data are required. However, labeling is high cost task and if automated, there is no guarantee that the label will always be correct. In the case of supervised learning in such a noisy labels environment, the accuracy of the model increases at the initial stage of learning, but decrease significantly after a certain period of time. There are various methods to solve the noisy label problem. But in most cases, the probability predicted by the model is used as the pseudo label. So, we proposed a method to predict the true label more quickly by refining the probabilities predicted by the model. Result of experiments on the same environment and dataset, it was confirmed that the performance improved and converged faster. Through this, it can be applied to methods that use the probability distribution predicted by the model among existing studies. And it is possible to reduce the time required for learning because it can converge faster in the same environment.