• 한국어정보학회:학술대회논문집
• /
• 2016.10a
• /
• pp.189-194
• /
• 2016

본 연구는 다국어 감성사전 및 감성주석 코퍼스 구축 프로젝트인 MUSE 프로젝트의 일환으로 한국어 감성사전을 구축하기 위해 대표적인 영어 감성사전인 SentiWordNet을 이용하여 한국어 감성사전을 구축하는 방법의 의의와 한계점을 검토하는 것을 목적으로 한다. 우선 영어 SentiWordNet의 117,659개의 어휘중에서 긍정/부정 0.5 스코어 이상의 어휘를 추출하여 구글 번역기를 이용해 자동 번역하는 작업을 실시하였다. 그 중에서 번역이 되지 않거나, 중복되는 경우를 제거하고, 언어학 전문가들의 수작업으로 분류해낸 결과 3,665개의 감성어휘를 획득할 수 있었다. 그러나 이마저도 병명이나 순수 감성어휘로 보기 어려운 사례들이 상당수 포함되어 있어 실제 이를 코퍼스에 적용하여 감성어휘를 자동 판별했을 때에 맛집 코퍼스에서의 재현율(recall)이 긍정과 부정에서 각각 47.4%, 37.7%, IT 코퍼스에서 각각 55.2%, 32.4%에 불과하였다. 이와 더불어 F-measure의 경우, 맛집 코퍼스에서는 긍정과 부정의 값이 각각 62.3%, 38.5%였고, IT 코퍼스에서는 각각 65.5%, 44.6%의 낮은 수치를 보여주고 있어, SentiWordNet 기반의 감성사전은 감성사전으로서의 역할을 수행하기에 충분하지 않은 것으로 나타났다. 이를 통해 한국어 감성사전을 구축할 때에는 한국어의 언어적 속성을 고려한 체계적인 접근이 필요함을 역설하고, 현재 한국어 전자사전 DECO에 기반을 두어 보완 확장중인 SELEX 감성사전에 대해 소개한다.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.26 no.1
• /
• pp.8-14
• /
• 2020

According to the statistics provided by Korean Maritime Safety Tribunal for the year 2018, the majority of marine accidents in the past four years have occurred in fishing boats and small-sized boats. Negligent behavior resulting from not looking outside and non-compliance with navigation laws are the primary reasons behind ship collisions. Although safety education and training are reinforced to prevent such accidents, they still occur frequently. Hence, technical methods are continuously being developed to reduce ship collisions caused by human cause. The objective of this study is to reduce ship collisions by employing the WAVE communication system, which has short transmitting and receiving periods that can be incorporated for high-speed small-sized vessels. In this study, the suitability of the communication range was examined, and the appropriate range and timing for avoidance motion were accordingly selected, and a control algorithm based on the same was thereby designed. Consequently normal operation of the collision avoidance system was verified by connecting and simulating the proposed WAVE communication router-controller-steering equipment.

• Membrane Journal
• /
• v.28 no.4
• /
• pp.271-278
• /
• 2018

In this study, the carbon electrodes using activated carbon fibers (ACFs) were prepared for the capacitive deionization process. The Polyvinylidene fluoride (PVDF) was used as the binder and the mixed ACFs with proper solvent was cast on the commercial graphite sheets to prepare the carbon electrodes. At this moment, the different particle sizes of ACFs were applied and the mixing ratio of solvent, PVDF and ACFs, 80 : 2 : 18 and 80 : 5 : 15, were used for the electrode preparation. Then their salt removal efficiencies were characterized under the various operating conditions, adsorption potential and time, desorption potential and time, concentration of feed NaCl solution and flow rate as well. Typically, the salt removal efficiency of 53.6% were obtained at the particle size below $32{\mu}m$, mixing ratio 80 : 2 : 18, adsorption 1.2 V and 3 min, desorption -0.1V and 1 min, and 15 mL/min flow rate of NaCl 100 mg/L.

• Korean Chemical Engineering Research
• /
• v.55 no.4
• /
• pp.561-566
• /
• 2017

Butt-fusion welding process is used to join the polymeric pipes. Recently, some researchers suggest the curved surface to enhance a welding quality. We investigated how curved welding surface affects heat and flow behaviors of polymer melt during the process in 2D axisymmetric domain with finite element method, and discussed the effect to the welding quality. In this study, we considered HDPE pipes. In heat soak stage, curved phase interface between the melt and solid is shown along the shape of welding surface. In jointing stage, squeezing flow is generated between curved welding surface and phase interface. The low shear rate in fusion domain reduces the alignment of polymer to the perpendicular direction of pipes, and then this phenomenon is expected to help to enhance the welding quality.

• Journal of Korean Society of Transportation
• /
• v.25 no.2 s.95
• /
• pp.95-105
• /
• 2007

Speeding and drunken driving make drivers fail to detect hazards and cope with various driving situations. These behaviors also raise the possibility of being involved in traffic accidents and tend to increase the number of fatalities. The authors compared the driving behavior determinants of a rockless drivers group, consisting of individuals who have committed traffic accidents or offended regulations through drunken driving, with a normal drivers group. In the results, the reckless drivers group showed high scores of 'speeding' and 'drunken driving', and they also stated that they had more experiences of speeding, drunken driving and traffic accidents. In the path analysis study, it was found that the impacts of the rockless drivers group's 'risk sensitivity' and 'situational adaptability' on traffic accidents were stronger than those of normal drivers. This means 'risk sensitivity' and 'situational adaptability' can explain the origins of traffic accidents better in the reckless drivers group than accidents of the normal drivers group.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.189-194
• /
• 2016

본 연구는 다국어 감성사전 및 감성주석 코퍼스 구축 프로젝트인 MUSE 프로젝트의 일환으로 한국어 감성사전을 구축하기 위해 대표적인 영어 감성사전인 SentiWordNet을 이용하여 한국어 감성사전을 구축하는 방법의 의의와 한계점을 검토하는 것을 목적으로 한다. 우선 영어 SentiWordNet의 117,659개의 어휘중에서 긍정/부정 0.5 스코어 이상의 어휘를 추출하여 구글 번역기를 이용해 자동 번역하는 작업을 실시하였다. 그 중에서 번역이 되지 않거나, 중복되는 경우를 제거하고, 언어학 전문가들의 수작업으로 분류해 낸 결과 3,665개의 감성어휘를 획득할 수 있었다. 그러나 이마저도 병명이나 순수 감성어휘로 보기 어려운 사례들이 상당수 포함되어 있어 실제 이를 코퍼스에 적용하여 감성어휘를 자동 판별했을 때에 맛집 코퍼스에서의 재현율(recall)이 긍정과 부정에서 각각 47.4%, 37.7%, IT 코퍼스에서 각각 55.2%, 32.4%에 불과하였다. 이와 더불어 F-measure의 경우, 맛집 코퍼스에서는 긍정과 부정의 값이 각각 62.3%, 38.5%였고, IT 코퍼스에서는 각각 65.5%, 44.6%의 낮은 수치를 보여주고 있어, SentiWordNet 기반의 감성사전은 감성사전으로서의 역할을 수행하기에 충분하지 않은 것으로 나타났다. 이를 통해 한국어 감성사전을 구축할 때에는 한국어의 언어적 속성을 고려한 체계적인 접근이 필요함을 역설하고, 현재 한국어 전자사전 DECO에 기반을 두어 보완 확장중인 SELEX 감성사전에 대해 소개한다.

• Journal of the Korean Association of Oral and Maxillofacial Surgeons
• /
• v.31 no.6
• /
• pp.474-480
• /
• 2005

We evaluated the need for prophylactic postoperative oral antibiotic medication in extraction of asymptomatic impacted mandibular third molars. All patient didn't show sign of pain, inflammation, swelling and trismus at the time of extraction. In the experimental group, oral antibiotic medication(Amoxicillin) was carried out for 5 days postoperatively. In the control group, the patients received no antibiotic medication. All groups didn't use antibiotic irrigation solution. Rule of group composition randomized. The surgical technique was the same in all cases. Parameters that were evaluated were infection, pain, facial swelling, trismus. We could not find any significant difference between the experimental and control groups.(P<0.05) The results of our study show that post operative oral prophylactic antibiotic medication after the extraction of impacted mandibular third molars does not contribute to less infection, pain, facial swelling and increased mouth opening after surgery. Therefore we suggest that prophylactic postoperative oral antibiotic medication is not needed in extraction of asymptomatic impacted mandibular third molars.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.13 no.8
• /
• pp.3806-3813
• /
• 2012

The coastal environment is affected by Cu and Zn contamination through diverse routes, including antifouling agents, industrial wastewater, and domestic activities. In this study, Crassostrea gigas and Littorina brevicula were used to examine the distributions of Cu and Zn, which are related to the increase in the use of new antifouling agents that have been used for ships since the TBT ban in 2003. The average Cu and Zn concentrations in the samples were 231.8 ${\mu}g/g$ and 492.6 ${\mu}g/g$, respectively. Both the Cu and Zn concentrations were highest in the shipyard samples, followed by the harbor sample for Cu and the industrial area sample for Zn. The results indicated that the Cu concentration was affected by the antifouling agents, and the Zn concentration was affected by both the antifouling agents and industrial wastewater. The areas that need intensive control to address the Cu and Zn contamination in major Korean coasts were identified.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.4
• /
• pp.381-387
• /
• 2019

The limitations of applying a variety of artificial intelligence to the medical community are, first, subjective views, extensive interpreters and physical fatigue in interpreting the image of an interpreter's illness. And there are questions about how long it takes to collect annotated data sets for each illness and whether to get sufficient training data without compromising the performance of the developed deep learning algorithm. In this paper, when collecting basic images based on acne data sets, the selection criteria and collection procedures are described, and a model is proposed to classify data into small loss rates (5.46%) and high accuracy (96.26%) in the sequential structure. The performance of the proposed model is compared and verified through a comparative experiment with the model provided by Keras. Similar phenomena are expected to be applied to the field of medical and skin care by applying them to the acne classification model proposed in this paper in the future.

• Korean Journal of Culture and Social Issue
• /
• v.18 no.3
• /
• pp.415-433
• /
• 2012

This study was conducted in order to explain the effect of driving behavior determinants such as drivers' personality and attitude that may induce risky driving behavior and to develop a valid method for discriminating risky drivers using the determinants. In the results of surveying 534 adult drivers, 5 driving behavior determinants (avoidance of problems, benefit/stimulus seeking, interpersonal anxiety, interpersonal anger, and aggression) were found to have a statistically significant effect on drivers' various risky driving behaviors. Using these factors, drivers were grouped according to risk levels (normal drivers, unintentionally risky drivers, and intentionally risky drivers). This result suggests that drivers' dangerous behavior level can be predicted using psychological factors such as their personality and attitude. Accordingly, if the driving behavior determinant model and the base score system used in this study are improved through further research, they are expected to be useful in predicting drivers' recklessness in advance, identifying problems, and providing differentiated safe driving education services based on the results.