• 한국콘텐츠학회논문지
• /
• 제21권2호
• /
• pp.121-129
• /
• 2021

본 연구에서는 데이터마이닝(Data Mining) 기법 중 하나인 연관관계분석(Association Rule Mining)을 적용하여 위험화물 선별모델을 구축함으로써 관세위험을 최소화하고자 한다. 이를 위해 관세청 수입신고서 빅데이터를 활용하여 연관관계분석 알고리즘인 어프라이어리 알고리즘(Apriori Algorithm)을 적용하고 공급망 간의 위험정도를 계산한다. 대규모의 수입신고 데이터로부터 해외공급자와 수입업체 간의 세율관련(과세가격, 품목, 중수량 등), 원산지표시 위반 등에 관련한 적발결과 관한 규칙셋(Rule Set)과 이 규칙들의 신뢰도(Confidence)을 확보하여 우범공급망 간의 거래패턴을 예측할 수 있는 선별모델을 구축한다. 총 2년 6개월 치의 수입신고 데이터를 활용하여 5-겹 교차검증(5-fold cross validation)을 수행한 결과 16.6%의 Precision과 33.8%의 Recall을 보였다. 이는 빈도기반 방법보다 Precision 기준 약 3.4배 Recall 기준 약 1.5배 높은 결과이다. 이로써 논문에서 제안하고 있는 방법이 관세위험을 줄일 수 있는 효과적인 방법임을 확인하였다.

• 전기학회논문지
• /
• 제67권11호
• /
• pp.1455-1462
• /
• 2018

As the interests in eco-friendly energy has increased, the interests in Electric Vehicles(EVs) are increasing as well. Moreover, due to the government's economic support for EVs, penetration level of it has rapidly increased. These sharp increases, however, induce various problems in distribution system, such as voltage/frequency variations, peak demand increasement, demand control, etc. To minimize these possible matters, lots of research have conducted. Nevertheless, most of it assumed extremely important factors, such as numbers and charging patterns of EVs. It inevitably results in errors in their research, and thus make it difficult to prevent the possible matters from EVs. In this paper, therefore, we use actual EVs charging data from KEPCO, and analysis and deduction of it were conducted. The simulations were carried out for four aspect(season, region, purpose).

• 한국멀티미디어학회논문지
• /
• 제25권2호
• /
• pp.237-246
• /
• 2022

Among the technologies of the 4th industrial revolution, drones that have grown rapidly and are being used in various industries can be operated by the pilot directly or can be operated automatically through programming. In order to be controlled by a pilot or to operate automatically, it is essential to predict and analyze the optimal path for the drone to move without obstacles. In this paper, after securing and analyzing the pilot training dataset through the unmanned aerial vehicle piloting training platform designed through prior research, the profile of the dataset that should be preceded to search and derive the optimal route of the unmanned aerial vehicle was designed. The drone pilot training data includes the speed, movement distance, and angle of the drone, and the data set is visualized to unify the properties showing the same pattern into one and preprocess the properties showing the outliers. It is expected that the proposed big data-based profile can be used to predict and analyze the optimal movement path of an unmanned aerial vehicle.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2020년도 추계학술대회
• /
• pp.54-55
• /
• 2020

4차 산업혁명으로 인공지능, 사물인터넷, 빅데이터 등의 기술이 조선 해운 산업에 매우 밀접하게 연관 되고 있고 이는 자율운항선박의 탄생을 가져왔다. 현재 선박의 기술적 특성상 속력을 갑자기 낮출 수 없으므로 항만에 접안하기 위해 예인선의 도움, 도선사의 승선, 육상관제센터의 선박 컨트롤 등 복잡한 커뮤니케이션을 필요로 한다. 본 연구에서는 자율운항선박이 도입될 경우 선박이 입항하기 위한 컨트롤 기준을 어떻게 설정할지 해결하고자 클러스터링 분석을 사용하였다. 입항 선박의 축적된 AIS 데이터를 기반으로 입항 패턴을 정량적으로 단계화하고자 K-Means 클러스터링을 사용했고 SOG(Speed over Ground), COG(Course over Ground), ROT(Rate of Turn)를 사용하여 입항 단계를 6개로 구분하였다.

• KIEAE Journal
• /
• 제14권5호
• /
• pp.59-65
• /
• 2014

Nowadays, reduction of energy use in buildings is a big issue, especially in public buildings like schools. The building structure is very simple in that, the room size, schedule and user number is similar across different schools. There are many policies which are suitable for this kind of buildings. Investigation of energy consumption pattern in primary school, middle school and high school in different cities of Korea has been done in this paper using statistical data from national organization and the data from IBM and Gyeonggi Provincial Office of Education, aimed at providing the basic data for the development of energy efficiency improvement policies of educational facilities. The study was divided according to climate, energy source type and public or private school, as different cities have different climates and accordingly different amount of energy sources are used. It was observed that, the average energy consumption in primary school is $36.9kWh/m^2$, in middle school is $20.5kWh/m^2$ and in high school $27.4kWh/m^2$. As further analysis, monthly energy consumption pattern has been analyzed for one city.

• 한국전자통신학회논문지
• /
• 제14권4호
• /
• pp.781-786
• /
• 2019

오늘날 다양한 생명현상을 다루고있는 질병연구와 같은 효율적인 목적을 달성하기 위해서는 이들 연구로부터 획득한 빅데이터를 처리하여 효과적인 현실적 가치를 부여할 수 있어야 한다. 본 논문에서 제안한 MKSV알고리즘은 최적의 확률분포를 추정하여 입력패턴을 결정 한 후 데이터마이닝 기법으로 분류한 결과 효율적인 계산량과 인식률을 획득할 수 있었다. MKSV 알고리즘은 유전자 데이터의 확률적 흐름을 시뮬레이션하여 빅데이터의 데이터마이닝 과정을 통해 데이터를 분류하여 빠르고 효과적인 성능 향상을 보임으로써 현 사회에 급증하는 질병과 유전자의 관련성을 연구하는 데 유용할 것이다.

• 한국과학교육학회지
• /
• 제37권6호
• /
• pp.1005-1014
• /
• 2017

본 연구는 대중의 과학적 소양을 함양하기 위한 기관으로서의 과학관의 역할에 주목하고, 우리나라 과학관이 대중에게 어느 정도 영향을 미치고 있는지 알아보기 위하여 소셜 빅데이터 분석을 통해 대중의 과학관에 대한 인식과 사용 행태를 알아보고자 하였다. 이를 위해 네이버 블로그와 트위터에에서 '과학관'이 포함된 게시글들을 추출한 뒤 텍스트 네트워크 분석, 빈도 분석, 공기어 분석 및 의미 분석을 실시하고 영어권의 분석 결과와 비교해 보았다. 그 결과 블로그에서는 주로 어린 자녀를 둔 부모 층에서 과학관이 이슈가 되고 있었고, 트위터에서는 단체 관람을 하는 학생 층이 다수 드러났다. 따라서 우리나라 대중들은 과학관을 주로 아이의 체험을 위한 공간으로 활용하고 있었고, 이 경우 과학관의 프로그램과 전시에 대해서는 긍정적으로 인식하고 있었다. 한편 단체 관람하는 학생들은 다소 부정적 감정을 표출하고 있는 것으로 나타났다. 과학관과 대중과의 소통, 대중의 과학에 대한 참여 등 제 3세대적 과학관의 기능적 측면에서 외국의 사례와 비교해본 결과 우리나라 대중들은 과학관 관람 이후 관람한 과학적 내용에 대한 언급이 거의 없었고, 논쟁이나 심포지움 등 과학적 의사소통과 관련된 언급 역시 거의 없었다. 또한 해설사나 직원들도 외국과는 달리 전혀 회자되지 않고 있었다. 한편, 영어권 게시글의 동사 분석에서 '배우다', '참여하다', '듣다', '읽다', '묻다', '생각하다', '그리다' 등의 유의미한 활동과 관련된 동사들이 다수 나타난 것에 비해 우리나라 게시글에서는 '물어보다', '생각하다' 가 소수 나타나는 것에 그치고 있었다. 따라서 과학관은 과학관 관람객들이 관람을 마친 뒤에 그들의 기억에 남고 대중들 사이에서 회자될 만큼 영향력 있고 다양한 내용과 활동이 일어날 수 있도록 개선할 필요가 있겠다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.1957-1980
• /
• 2021

The significant aim of this research has always been to evaluate the mechanism for efficient and inherently aware usage of vitality in-home devices, thus improving the information of smart metering systems with regard to the usage of selected homes and the time of use. Advances in information processing are commonly used to quantify gigantic building activity data steps to boost the activity efficiency of the building energy systems. Here, some smart data mining models are offered to measure, and predict the time series for energy in order to expose different ephemeral principles for using energy. Such considerations illustrate the use of machines in relation to time, such as day hour, time of day, week, month and year relationships within a family unit, which are key components in gathering and separating the effect of consumers behaviors in the use of energy and their pattern of energy prediction. It is necessary to determine the multiple relations through the usage of different appliances from simultaneous information flows. In comparison, specific relations among interval-based instances where multiple appliances use continue for certain duration are difficult to determine. In order to resolve these difficulties, an unsupervised energy time-series data clustering and a frequent pattern mining study as well as a deep learning technique for estimating energy use were presented. A broad test using true data sets that are rich in smart meter data were conducted. The exact results of the appliance designs that were recognized by the proposed model were filled out by Deep Convolutional Neural Networks (CNN) and Recurrent Neural Networks (LSTM and GRU) at each stage, with consolidated accuracy of 94.79%, 97.99%, 99.61%, for 25%, 50%, and 75%, respectively.

• International Journal of Contents
• /
• 제17권1호
• /
• pp.1-10
• /
• 2021

Visual analytics is an emerging research field that combines the strength of electronic data processing and human intuition-based social background knowledge. This study demonstrates useful visual analytics with Tableau in conjunction with semantic network analysis using examples of sentiment flow and strategic communication strategies via Twitter in a blockchain domain. We comparatively investigated the sentiment flow over time and language usage patterns between companies with a good reputation and firms with a poor reputation. In addition, this study explored the relations between reputation and marketing communication strategies. We found that cryptocurrency firms more actively produced information when there was an increased public demand and increased transactions and when the coins' prices were high. Emotional language strategies on social media did not affect cryptocurrencies' reputations. The pattern in semantic representations of keywords was similar between companies with a good reputation and firms with a poor reputation. However, the reputable firms communicated on a wide range of topics and used more culturally focused strategies, and took more advantages of social media marketing by expanding their outreach to other social media networks. The visual big data analytics provides insights into business intelligence that helps informed policies.

• 대한가정학회지
• /
• 제38권7호
• /
• pp.1-12
• /
• 2000

Although the term 'family business' is relatively new, this style of business is universal. An Unusual feature that must be noted, is that even though it is a common style of business is not clearly defined. The purpose of this study is to identify the different family participation patterns, and the variables that effect different types of participation. '1997 Daewoo Panel Data' was used. Some descriptive statistics and a multinomial logit model were employed for the analysis. The standard type of business focused on in this study was a family owned and operated 'ma and pa' typed business and the sample was limited to households where one or both of the partners involved in a family owned and operated business. The main resets obtained from this sample were as follows: 1. Personal characteristics such as respondents' gender, age and educational level were important variables that effected the participation of family members in the business. As can be seen in the gender analysis, family businesses owned by men showed all available patterns of family operated businesses in relatively high numbers. A large percentage of businesses owned women were of self-employed pattern. According to the analysis by age and educational level, young people with a high level of education tend to managed their small businesses by employing others rather than utilising the self-employed or family operated pattern. 2. While big families showed a high percentage of a combination pattern of a family-run, and ordinary employer/employee company, relatively small families usually opted for purely family-run businesses. Whether the family have children under 6 or not, and the number of children under 6 did net significantly effect to the patterns of the family system of small family businesses. 3. The size, location and kind of family business also effected participation patterns of the family members significantly. These results suggest that further study will be required to gain more exact and meaningful information to help Korean family businesses.