• 한국항만경제학회지
• /
• 제39권2호
• /
• pp.1-13
• /
• 2023

이 연구는 IMO의 환경규제와 4차산업 혁명 기술의 확산에 따라 그중요성과 비중이 확대되고 있는 친환경 스마트 선박의 성장에 필요한 인력 수요를 통계청의 2000년~2020년의 조선산업 인력자료를 기반으로 예측하였다. 추세분석과 시계열분석의 다양한 모델을 적용하여 조선산업의 인력 수요를 예측하고 최근 5년간의 실적치와 비교하여 기하평균을 적용한 단순평균법이 예측 오차가 유의적으로 가장 적은 것으로 평가되었다. 그리고 산업통상자원부의 친환경 스마트 선박 분야의 2018년과 2020년의 인력현황 설문조사 결과를 바탕으로 조선산업 인력 증가추이를 반영하여 인력 수요를 예측하였다. 조선산업의 인력수요 예측치에 친환경 스마트 선박부분의 인력 증가수치를 반영하여 인력 수요를 예측한 결과, 2025년 62,001명, 2030년 85,035명으로 증가하는 것으로 예측되었다. 본 연구는 고부가가치 친환경 스마트 선박 분야에 필요한 인력 수요를 통계자료에 기반하여 객관적으로 예측함으로써, 향후의 인력 수요에 대응한 적절한 전문인력의 양성 및 공급 방안 수립에 기여하게 될 것으로 평가된다.

• Journal of Information Processing Systems
• /
• 제15권3호
• /
• pp.694-706
• /
• 2019

In this study, we applied the long short-term memory (LSTM) model to classify the cryptocurrency price time series. We collected historic cryptocurrency price time series data and preprocessed them in order to make them clean for use as train and target data. After such preprocessing, the price time series data were systematically encoded into the three-dimensional price tensor representing the past price changes of cryptocurrencies. We also presented our LSTM model structure as well as how to use such price tensor as input data of the LSTM model. In particular, a grid search-based k-fold cross-validation technique was applied to find the most suitable LSTM model parameters. Lastly, through the comparison of the f1-score values, our study showed that the LSTM model outperforms the gradient boosting model, a general machine learning model known to have relatively good prediction performance, for the time series classification of the cryptocurrency price trend. With the LSTM model, we got a performance improvement of about 7% compared to using the GB model.

• Journal of Information Technology Applications and Management
• /
• 제12권1호
• /
• pp.141-155
• /
• 2005

It is very important that sellers provide reasonable reserve prices for auction items in internet auction systems. Recently, an agent has been proposed to generate reserve prices automatically based on the case similarity of information retrieval theory and the moving average of time series analysis. However, one problem of the previous approaches is that the recent trend of auction prices is not well reflected on the generated reserve prices, because it simply provides the bid price of the most similar item or an average price of some similar items using the past auction data. In this paper. in order to overcome the problem. we propose a method that generates reserve prices based on the moving average. the exponential smoothing, and the least square of time series analysis. Through performance experiments. we show that the successful bid rate of the new method can be increased by preventing sellers from making unreasonable reserve prices compared with the previous methods.

• 한국콘텐츠학회논문지
• /
• 제16권6호
• /
• pp.293-304
• /
• 2016

이 연구는 국내외 모션 인포그래픽 공모전 수상작을 대상으로 디자인 트렌드를 분석한 것이다. 정보디자인 측면에서 모션 인포그래픽 디자인 과정을 정보 구조화, 정보 시각화로 나누고 각 단계에서 고려되어야할 디자인 요소들을 추출하여 각 기준에 맞추어 작품들을 분석하였다. 분석결과 국내외 작품에서 함께 나타난 공통점들은 모션 인포그래픽이란 매체를 통한 커뮤니케이션이 어떤 방식으로 이루어지며 그 방식들의 효용성에 대해 이해할 수 있는 시사점을 주었다. 동시에 차이점을 통해서는 국내외의 모션 인포그래픽에 대한 인식의 차이점을 확인할 수 있었으며 국내 모션 인포그래픽의 다양성 부재, 표현의 획일화 등을 읽어낼 수 있었다. 본 연구의 분석결과는 모션 인포그래픽을 제작하는 실무와 교육 분야에서 디자인 전략 수립의 틀로 그 유용성을 제공할 것으로 본다.

• 한국문헌정보학회지
• /
• 제38권4호
• /
• pp.105-123
• /
• 2004

본 연구는 도서관${\cdot}$정보센터가 미래의 변화에 능동적으로 대처해야 한다는 전제아래 미래예측방법의 하나인 추세연상적 예측을 통하여 도서관${\cdot}$정보센터의 미래를 예측한 것이다. 전술한 연구의 목적을 달성하기 위하여 문헌연구와 주요 도서관의 홈페이지 분석, 전문사서와의 인터뷰를 실시하고 그 결과, 나타난 추세를 근거로 도서관${\cdot}$정보센터의 중기 미래를 예측하였다. 본 연구의 결과, 얻어진 결론은 다음과 같다. 첫째, 미래의 정보환경은 모바일과 유비쿼터스 환경으로 대표되며, 이들을 통해서 이용자는 언제, 어디서나, 어떠한 기기를 이용하더라도 다양한 미디어에 접근할 수 있게 된다. 둘째, 정보기술의 발전에 따른 시간적 지체 단말기 기술과 보급속도 도서관 서비스의 자세와 대응 속도 저작권 문제 등이 미래의 도서관에 영향을 미치는 불확실 요인이다. 셋째, 미래의 도서관은 무선 단말기를 통한 정보의 접근과 이용이 더욱 확대되며, 모바일 기술이 발전하고 유비쿼터스 환경이 보편화되어, 결국에는 전자 정보의 이용이 더욱 늘어난다. 넷째 미래의 도서관은 끊임없이 변하는 정보기술 환경에 대처하기 위해서 이용자의 요구와 그에 따른 정보서비스의 추이를 분석하고 예측하여 다양한 상황에 적절히 대처할 수 있는 전략을 마련하여야 한다.

• Journal of Astronomy and Space Sciences
• /
• 제27권3호
• /
• pp.231-238
• /
• 2010

We analyzed global positioning system (GPS)-derived precipitable water vapor (PWV) trends of the Korea Astronomy and Space Science Institute 5 stations (Seoul, Daejeon, Mokpo, Milyang, Sokcho) where Korea Meteorological Administration meteorological data can be obtained at the same place. In the least squares analysis, the GPS PWV time series showed consistent positive trends (0.11 mm/year) over South Korea from 2000 to 2009. The annual increase of GPS PWV was comparable with the 0.17 mm/year and 0.02 mm/year from the National Center for Atmospheric Research Earth Observing Laboratory and Atmospheric InfraRed Sounder, respectively. For seasonal analysis, the increasing tendency was found by 0.05 mm/year, 0.16 mm/year, 0.04 mm/year in spring (March-May), summer (June-August) and winter (December-February), respectively. However, a negative trend (-0.14 mm/year) was seen in autumn (September-November). We examined the relationship between GPS PWV and temperature which is the one of the climatic elements. Two elements trends increased during the same period and the correlation coefficient was about 0.8. Also, we found the temperature rise has increased more GPS PWV and observed a stronger positive trend in summer than in winter. This is characterized by hot humid summer and cold dry winter of Korea climate and depending on the amount of water vapor the air contains at a certain temperature. In addition, it is assumed that GPS PWV positive trend is caused by increasing amount of saturated water vapor due to temperature rise in the Korean Peninsula. In the future, we plan to verify GPS PWV effectiveness as a tool to monitor changes in precipitable water through cause analysis of seasonal trends and indepth/long-term comparative analysis between GPS PWV and other climatic elements.

• 한국농공학회논문집
• /
• 제66권2호
• /
• pp.25-34
• /
• 2024

The long-term monitoring of the Soyang Lake's water quality, covering 25% of the North Han River watershed, is crucial for effective management of both lake water quality and pollution sources in the broader region. This study utilized continuous monitoring data from the front of the Soyang Dam spanning 2003 to 2022, aiming to analyze trends and provide foundational insights for water quality management. Results revealed a slightly poor grade (IV) for total nitrogen (T-N) in both surface and mid-depth layers, indicating a need for concentrated T-N management. Trend analyses using the Mann-Kendall test and Sen's Slope depicted a decreasing trend in total phosphorus (T-P) for both layers, attributed to non-point source pollution reduction projects initiated after the Soyang Lake's designation as a pollution control area in 2007. The LOWESS analysis showed a T-P increase until 2006, followed by a decrease, influenced by the impact of Typhoon Ewiniar in that year. This 20-year overview establishes a comprehensive understanding of the Soyang Lake's water quality and trends, allowing for a seasonal and periodical analysis of water quality changes. The findings underscore the importance of continued monitoring and management strategies to address evolving water quality issues in the Soyang Lake over time.

• The International Journal of Costume Culture
• /
• 제4권3호
• /
• pp.229-240
• /
• 2001

The purpose of the present study was to analyze the acceptance level of forecasted information of casual wear in tate 1990s in Korea and the way of utilizing fashion trends information by casual wear industries. The Present study was implemented by content analysis and descriptive survey using questionnaire and interview. Trends information in fashion journals published by fashion institute and articles in daily newspapers were analyzed in terms of fashion image, color, fabric, and silhouette. The data collected from questionnaire and interview with 113 fashion specialists were analyzed through frequency, percentage. The results indicated that among the forecasted information regarding fashion image, romantic and feminine images showed a high level of acceptance to national brand women's casual wear in the late 1990s, while mannish image showed a low level of acceptance. For men's casual wear in the same time period, androgynous trends appeared most frequently, not only in forecasted information, but also in actual trend. it was forecasted that yellow, white, and gray would be in trend and those colors appeared frequently in actual trend. On the other hand pastel tone appeared much more frequently than forecasted. Natural, thin - transparent (S/S) and stretch fabrics (F/W) were in actual trend as it was forecasted. Fit and Pare (woman), and long and slim (man) silhouettes were in actual trend as if was forecasted, but barrel silhouette appeared only in forecasted information. Most of the information forecasting fashion trends for next season were applied to the product planning of the season, right after the information comes out.

• 한국추진공학회지
• /
• 제12권6호
• /
• pp.1-6
• /
• 2008

본 연구에서는 계측 데이터의 성능 추이를 분석하여 가스터빈 엔진의 결함 여부를 탐지하기 위한 퍼지 경향감시 방법을 제안하였다. 제안된 경향감시 방법은 연료유량, 배기가스 온도, 로터회전수, 진동수와 같은 중요 엔진 파라미터를 모니터링 하여 시간에 따른 변화를 분석하여 엔진 상태를 진단하는 것이다. 이를 위해 먼저 선형회귀분석을 통해 엔진 상태 변화를 수식화하고 퍼지 로직을 통해 진단 결과를 분석하여 예측되는 손상 원인을 제시한다.

• International Journal of Aeronautical and Space Sciences
• /
• 제9권2호
• /
• pp.64-70
• /
• 2008

The UAV propulsion system that will be operated for long time at more than 40,000ft altitude should have not only fuel flow minimization but also high reliability and durability. If this UAV propulsion system may have faults, it is not easy to recover the system from the abnormal, and hence an accurate diagnostic technology must be needed to keep the operational reliability. For this purpose, the development of the health monitoring system which can monitor remotely the engine condition should be required. In this study, a fuzzy trend monitoring method for detecting the engine faults including mechanical faults was proposed through analyzing performance trends of measurement data. The trend monitoring is an engine conditioning method which can find engine faults by monitoring important measuring parameters such as fuel flow, exhaust gas temperatures, rotational speeds, vibration and etc. Using engine condition database as an input to be generated by linear regression analysis of real engine instrument data, an application of the fuzzy logic in diagnostics estimated the cause of fault in each component. According to study results. it was confirmed that the proposed trend monitoring method can improve reliability and durability of the propulsion system for a long endurance UAV to be operated at medium altitude.