A5103220766- Advanced Text Analysis Techniques
- Computational and Text Analysis Methods
- Sentiment Analysis and Opinion Mining
- Text and Document Classification Technologies
- Marine and Coastal Research
- Natural Language Processing Techniques
- Topic Modeling
- Soil Moisture and Remote Sensing
- Language, Metaphor, and Cognition
- Water Quality Monitoring and Analysis
- Fault Detection and Control Systems
- Fire Detection and Safety Systems
- Text Readability and Simplification
- Diverse Approaches in Healthcare and Education Studies
- Discourse Analysis in Language Studies
- Infrared Target Detection Methodologies
- Folklore, Mythology, and Literature Studies
- Water Quality Monitoring Technologies
- Seismic Waves and Analysis
- Stonefly species taxonomy and ecology
- Video Surveillance and Tracking Methods
- Data Quality and Management
- Language, Discourse, Communication Strategies
- Oil Spill Detection and Mitigation
- UAV Applications and Optimization
Korea Maritime and Ocean University
2021-2022
The number of ship accidents occurring in the Korean ocean has been steadily increasing year by year. Korea Maritime Safety Tribunal (KMST) published verdicts to ensure that relevant personnel can share judgment on these accidents. As 2020, there have 3156 accidents; thus, it is difficult for study various only reading verdicts. Therefore, this study, we propose a multi-task deep learning model with an attention mechanism predicting sentencing tasks are accident types, applied articles, and...
Aspect-based sentiment analysis is a text technique that categorizes data by aspect and identifies the attributed to each one task for fine-grained analysis. In order accurately perform analysis, word within text, target it modifies, holder who represents are required; however, they should be extracted in sequence because an important clue extracting target, which key evidence of holder. Namely, three types information sequentially become clue. Therefore, this paper, we propose stepwise...
Metaphor and metonymy are essential linguistic devices in Maritime English, enabling the communication of complex ideas evoking cultural associations. Despite their importance, few studies have investigated role these maritime contexts. This study used a method that combines Python coding with manual identification to explore metaphorical metonymic expressions self-constructed corpus English. Specifically, we developed codes based on previous research initially identify metaphors metonymies...
한국어 의존구조를 분석하는 데에는 몇 가지 고질적인 문제가 있다. 그 중 하나는 중심어 위치문제이고 다른 구성성분의 단위 문제이다. 이와 같은 문제는 구묶음을 수행함으로써 어느 정도는 해결된다. 구묶음은 형태소 분석과 구문분석의 중간 단계에 위치하면서 말덩이라 하는 구성성분을 찾는 과정이다. 본 논문에서는 말덩이의 정의와 의의를 살펴보고 말덩이 부착 말뭉치를 구축한다. 또한 구축된 말뭉치와 Bi-LSTM/CRFs를 이용한 제안한다. 실험을 통해서 제안된 구묶음 모델은 98.54%의 F1점수를 보여 실용적으로 사용할 수 있을 것으로 판단된다. 다양한 입력 표상에 따른 성능을 분석하여 fastText가 가장 좋은 보였다. 오류 분석을 통해 시스템의 문제를 향후 시스템 개선에 적극 활용할 계획이다.