• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.2
• /
• pp.308-313
• /
• 2021

In this study, tweets about the 2019 Koreas-US DMZ Summit were collected along with a time sequence and analyzed by a sequential topic modeling method, Dynamic Topic Modeling(DTM). In microblogging services such as Twitter, unstructured data that mixes news and an opinion about a single event occurs at the same time on a large scale, and information and reactions are produced in the same message format. Therefore, to grasp a topic trend, the contextual meaning can be found only by performing pattern analysis reflecting the characteristics of sequential data. As a result of calculating the DTM after obtaining the topic coherence score and evaluating the Latent Dirichlet Allocation(LDA), 30 topics related to news reports and opinions were derived, and the probability of occurrence of each topic and keywords were dynamically evolving. In conclusion, the study found that DTM is a suitable model for analyzing the trend of integrated topics in a specific event over time.

• Journal of the Korea Convergence Society
• /
• v.13 no.2
• /
• pp.13-20
• /
• 2022

This study is to develop a named entity recognition model specialized in criminal investigation domains using deep learning techniques. Through this study, we propose a system that can contribute to analysis of crime for prevention and investigation using data analysis techniques in the future by automatically extracting and categorizing crime-related information from text-based data such as criminal judgments and investigation documents. For this study, the criminal investigation domain text was collected and the required entity name was newly defined from the perspective of criminal analysis. In addition, the proposed model applying KoELECTRA, a pre-trained language model that has recently shown high performance in natural language processing, shows performance of micro average(referred to as micro avg) F1-score 98% and macro average(referred to as macro avg) F1-score 95% in 9 main categories of crime domain NER experiment data, and micro avg F1-score 98% and macro avg F1-score 62% in 56 sub categories. The proposed model is analyzed from the perspective of future improvement and utilization.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2022.05a
• /
• pp.273-273
• /
• 2022

혼합층(Mixed layer)은 온도가 일정한 수심층으로, 해수표면에 작용하는 바람의 영향으로 인하여 해수가 위아래로 섞여 형성된다. 이러한 혼합층은 영양염의 순환과 산소의 공급 등과 함께 일차생산량을 결정하는 중요한 요인이 될 수 있으며 혼합층 두께의 변동은 양식 산업에 영향을 미칠 수 있다. 최근에는 기후변화로 인한 해수면 상승 및 해수온 상승 등이 지속되고 있으며, 이러한 현상은 해양생태계의 변화를 초래하여 수산업의 피해를 유발할 수 있다(강원연구원, 2017). 이에 국립수산과학원, 기상청, 국립해양조사원 등 유관기관에서는 정선해양 수온 관측 및 해수순환모델을 이용하여 혼합층의 분석을 수행하고 있으나 격자 구축 및 초기·경계장 설정의 한계가 존재하여 정밀하고 정확한 혼합층 분석에는 어려움이 있다. 이에 본 연구에서는 비정형격자를 사용하여 격자 구축에 제약이 없는 SCHISM (Semi-implicit Cross-scale Hydroscience Integrated System Model)을 이용하여 우리나라 연안해역의 계절변화 및 기후변동성에 따른 혼합층 두께의 변화를 검토하고자 한다. 연구대상지는 서해·동해·남해를 포함한 우리나라 전체 연안 해역(위도: 32°N ~ 39°N, 경도: 124°E ~ 132°E)으로 선정하였으며, 격자크기 100 ~ 3,000 m인 삼각격자로 격자를 구축하였다. 혼합층을 분석하기 위하여 수직격자 층은 50층으로 SZ(Sigma Z coordinate system)좌표계를 사용하였다. 초기·경계장은 FES(Finite Element Solution)2014, HYCOM(Hybrid Coordinate Ocean Model) 및 대기모델 결과를 이용하여 설정하였다. 수치모형 검증을 위하여 수온관측소에서 수심별 측정한 수온 값과 SCHISM 결과 값을 비교하였고, 상대오차가 약 10% 이내로 나타나 모형의 정확도를 확인하였다. 최종적으로 해수면 상승 및 해수온 상승 시나리오를 고려하여 계절별 연안해역의 혼합층 두께의 변화 양상에 대하여 검토하였다. 향후에는 보다 정밀한 대기모델과의 혼합모형 구축 및 다양한 수심 별 관측자료를 활용한다면 실무에서 적용 가능한 혼합층 분석 및 수산업 피해 발생 지역에 대한 피해저감 대책 수립이 가능할 것으로 판단된다.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.1
• /
• pp.609-615
• /
• 2024

The construction industry's fatality count stands at 402, comprising approximately 46% of total industrial accidents. Notably, construction costs less than 5 billion won account for about 69%, so strengthening safety management at small and medium-sized construction sites is required. In this study, 19,511 accident investigation data were collected using web scraping. Through statistical analysis of the collected structured data and text mining analysis of the unstructured data, accident types and causes of accidents were analyzed by construction costs at sites less than 5 billion won. As a result, it was confirmed that there were differences in accident types and causes depending on the construction costs. It is hoped that the results of this study will be used for customized safety management at small and medium-sized construction sites.

• Journal of Advanced Navigation Technology
• /
• v.16 no.2
• /
• pp.247-254
• /
• 2012

In this paper, we propose a concatenated diversity system to assure the data transmission reliability between flight type air nodes which move according to their atypical orbit, then its performance is analyzed using computer simulation and it is designed with hdl. The proposed system cannot only improve a bandwidth efficient and coding gain from diversity TCM code but also the reliability of data transmission is high. From the computer simulation result about bit error rate(BER) of the proposed system, we confirm that its BER performance is about 11dB greater than TCM code at $10^{-2}$ and about 11dB greater than space time block code at $10^{-3}$ which has a full diversity gain. In addition, when we compare its BER performance with space time trellis code which has both a diversity gain and a coding gain, the performance of the proposed system is greater than about 1.5dB at $10^{-5}$. Lastly, after designing the proposed system with HDL, we can confirm that the operation result is correct.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.4
• /
• pp.509-517
• /
• 2020

This study evaluated the importance of technical characteristics in manufacturing street furniture using 3D printing technology to suggest the direction of development of high priority 3D printing technology. The importance was analyzed by the QFD, quantified by scores and the priority of the items was summarized. As a result, the 'output size', 'shrinkage of material', and 'output angle' were derived as technical elements that should be prioritized in development and research. For verification, the design of atypical street furniture was made into a large 3D printed output and the development direction was suggested by applying the technical elements of priority during the manufacturing process. Street furniture should be designed based on functionality and stability, as well as economic efficiency, productivity, and aesthetics. Therefore, the 'output size' ensured stability by minimizing the division of parts, and the 'contractability of materials' satisfied the aesthetic and productivity by minimizing the error of form. Finally, the 'output angle' was verified by improving the quality of the output and selecting an angle with efficient and structural stability through various output angles.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.24 no.1
• /
• pp.35-42
• /
• 2020

Torsional behavior due to the plane irregularities of the piloti building can cause excessive story drift in the torsionally outermost column, which can lead to shear failure of the column. As a seismic retrofit method that can control the torsional behavior of the piloti building, the expansion of RC wall, steel frame or steel brace may be used, but such methods may hinder the openness of the piloti floor. Therefore, in this study, linear dynamic analysis and nonlinear static analysis for piloti buildings retrofitted by knee brace were performed, and seismic performance evaluation and torsion control effect of knee brace were analyzed. The results showed that the shear force of the column increased when the piloti building retrofitted by knee brace, but it was effective in controlling the torsional deformation. In case of retrofit between knee brace and column by 30°, the shear force of the column increased less than that of 60°, and the lateral displacement of column was decreased in the order of □, ◯ and Ｈ in cross-section.

• The Journal of the Korea Contents Association
• /
• v.21 no.12
• /
• pp.400-407
• /
• 2021

Recently, mukbang and cookbang or cooking shows on TV and YouTube channels have increased, and the influences of these broadcasts on food consumption have been gradually increasing. There were several news articles on 'Baek Jong-won effect', in which the consumption of the agri-food Mr. Jong-won Baek mentioned on his broadcast soared, and even foods named after him are on the market. In this study, Mr. Jong-won Baek, who produces influential cooking contents through various media, was taken as a representative example. We evaluated if 'Baek Jong-won effect' exists on potato consumption, which Mr. Jong-won Baek broadcasted potato cooking recipes on TV and YouTube. After the potato recipe was broadcasted for the first time on the TV show called HomeFoodRescue, the differences in the amount of money to purchase potatoes before and after the broadcast were estimated by using the money amount to purchase data of Agri-food consumers panel and the difference-in-differences method at 6 time points (3, 6, 9, 12, 24, and 36 months). Among the time points analyzed, the potato purchases at post-broadcast were less than those at pre-broadcast. No results were observed suggesting the existence of 'Baek Jong-won effect' on potato consumption through HomeFoodRescue show in the study.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.3
• /
• pp.745-753
• /
• 2010

Many studies using RFID which can identify the movement of targeted objects has been conducted actively. In this paper, we propose improved tagging method to check exactly the amount of fish feed by the aquaculture tank in a land-based aquaculture farm and design the RFID feeding system to perform auto feeding history. Also, we implement the prototype system which it includes development program for auto-identification and auto-recording and self-made RFID feeding cart equipped the weight device, RFID reader, control module and display device.

• Journal of Digital Convergence
• /
• v.11 no.8
• /
• pp.31-39
• /
• 2013

Recently, integration between social media and the industry has been expended, and as the usage of Internet through various smart devices of not only the existing PC but also smart phone, tablet PC and so on, a lot of unstructured data has occurred, leading to increased interest on big data system. According to the institutes which specialize in market research, the data amount is predicted to increase by 9 folds in the next 5 years when compared to the present, and the big data market is also expected to grow bigger. This dissertation evaluates the efficiency test of big data through analysis on the requirements by identifying and fragmenting the items of efficiency quality evaluation that big data should be equipped with.