• Proceedings of the Korea Water Resources Association Conference
• /
• 2020.06a
• /
• pp.416-416
• /
• 2020

우리나라의 계절 특성상 여름철 집중호우가 쏟아지는 현상이 빈번하게 발생하는데 이러한 돌발홍수가 예고 없이 일어나 상습적으로 침수 피해를 입는 지역이 증가하고 있다. 본 연구에서 2009년 ~ 2019년 동안 서울시 침수 피해 사건 중심의 인터넷 기사를 기반으로 실제 침수 사례를 조사해본 결과, 침수가 가장 많이 발생한 순으로 반포동(26건), 대치동(25건), 잠실동(21건)으로 집계되었다. 침수피해가 가장 많은 반포동을 연구지역으로 선정하고 그 중 잠수교의 수위를 예측하는 연구를 진행하였다. 기존 연구에서는 수치모형에 비해 신속한 결과를 도출할 수 있는 자료 기반 모형 중 LSTM 기법을 많이 사용하였다. 그러나 이는 선행 시간이 길어질수록 첨두수위에서 과소추정된 것으로 분석된 취약점이 존재하였다(정성호 외, 2018). 본 연구에서는 이러한 단점을 보완하기 위해 GANs(Generative Adversarial Networks)를 이용하였다. GANs는 생성자와 감별자가 나뉘어 생성자가 실제 자료인 첨두수위에서의 잠수교의 수위를 학습하고 실제와 근접한 가상데이터를 결과로 생성하여 감별자는 그 생성된 미래의 잠수교의 수위가 실제인지 가상인지 판별하도록 학습시키는 신경망 구조이다. 사용한 수문자료는 한강홍수통제소, 기상청, 국립해양조사원에서 제공하는 최근 15년간의 (2005년~2019년) 수위, 방류량, 강수량, 조위 자료를 수집하였고 t-test와 상관성분석을 통해 사용한 인자 간의 유의미성 판단과 상관성을 분석했다. 또한, 민감도 분석 결과 시퀀스길이(5), 반복횟수(1000), 은닉층(10), 학습률(0.005)로 최적값을 선정하였다. 또한 학습구간(2005년~2014년)과 검증구간(2015~2019년)으로 나누어 상대적으로 높은 수위가 관측되는 홍수기의 3, 6, 9시간 후의 수위를 예측하고 오차 지표를 이용해 평가하였다. LSTM 기법으로 예측된 수위와 GANs로 예측된 수위를 비교한 결과 GANs으로 예측된 첨두수위에서의 정확도가 5% 정도로 향상되었다. 향후에는 다양한 영향인자와 다른 기법과의 결합을 고려한다면 보다 정확하게 수위를 예측하여 하천 주변 사회기반시설의 침수 피해를 감소시킬 것으로 판단된다.

• The KIPS Transactions:PartD
• /
• v.13D no.6 s.109
• /
• pp.775-788
• /
• 2006

Many disaster monitoring systems are constantly studied to prevent disasters such as environmental pollution, the breaking of a tunnel and a building, flooding, storm earthquake according to the progress of wireless telecommunication, the miniaturization of terminal devices, and the spread of sensor network. A disaster monitoring system can extract information of a remote place, process sensor data with rules to recognize disaster situation, and provide work for preventing disaster. However existing monitoring systems are not enough to predict and prevent disaster, because they can only process current sensor data through utilizing simple aggregation function and operators. In this paper, we design and implement a disaster prevention system to predict near future dangerous area through using outside sensor network and spatial Information. The provided prediction technique considers the change of spatial information over time with current sensor data, and indicates the place that could be dangerous in near future. The system can recognize which place would be dangerous and prepare the disaster prevention. Therefore, damage of disaster and cost of recovery would be reduced. The provided disaster prevention system and prediction technique could be applied to various disaster prevention systems and be utilized for preventing disaster and reducing damages.

• Spatial Information Research
• /
• v.9 no.2
• /
• pp.239-261
• /
• 2001

A large suite of official statistical data sets has been compiled for geographical units under the national directives, and it is the quantitative regional analysis procedures that could add values to them. This paper reports our attempts at prototyping a statistical GIS which is capable of serving over the Web a variety of regional analysis routines as well as value-added statistics and maps. A pilot database of some major statistical data was ingested for the city of Seoul. The baseline subset of regional analysis methods of practical usage was selected and accommodated into the business logic of the target system, which ranges from descriptive statistics, regional structure/inequality measures, spatial ANOVA, spatial (auto) correlation to regression and residual analysis. The leading-edge information technologies including the application server were adopted in the system design and implementation so that the database, analysis modules and analytic mapping components may cooperate seamlessly behind the Web front-end. The prototyped system supports tables, maps, and files of downloadable format for input and output of the analyses. One of the most salient features of out proposed system is that both the database and analysis modules are extensible via the bi-directional interface for end users; The system provides users with operators and parsers for algebraic formulae such that the stored statistical variables may be transformed and combined into the newly-derived set of variables. This functionality eventually leads to on-the-fly fabrication of user-defined regional analysis algorithms. The stored dataset may also be temporarily augmented by user-uploaded dataset; The extension of this form, in essence, results in a virtual database which awaits for users commands as usual. An initial evaluation of the proposed system confirms that the issues involving the usage and dissemination of information can be addressed with success.

• Journal of Korea Technology Innovation Society
• /
• v.22 no.2
• /
• pp.207-219
• /
• 2019

The international trends in activities of government R&D performance measurement is evolving into evidence-based approach to support the verification of objective policy implementation. The European Commission has been implementing KETs (Key Enabling Technologies) policy that incubates nanotech based emerging technologies to support the fourth industrial revolution/revitalization of high-tech manufacturing, and resulting innovations are quantified by KETs Observatory project. The European Commission also built Innovation Radar system that monitors potentially innovative projects within FP7 and Horizon2020 by data analysis. The United States is also upgrading its Federal RePORTER system to quantitatively monitor federal R&D investments with outcomes (including nanotechnology). In the field of Korean nanotechnology, basic statistical data by analyzing NTIS (National Science & Technology Information Service) information is applied in policy field. Developing innovation measurement methodology beyond basic statistical analysis is an international policy issue, and a long-term R&D investment area of a government. The objective of this model study is to quantify the innovation potential of nano R&D investments conducted by Korea government, using input-output based efficiency measurement model and NTIS (National Science & Technology Information Service) that is comprehensive data portal for national R&D investments/outcomes including nanotechnology.

• Pediatric Infection and Vaccine
• /
• v.28 no.3
• /
• pp.144-148
• /
• 2021

Purpose: Nonpharmaceutical measures, such as social distancing, have resulted in unintended consequences, including a decrease in the incidence of childhood diseases. This study aimed to estimate the impact of social distancing on the incidence of febrile seizure (FS) in Korea using nationally representative data. Methods: We used claims data from the Health Insurance Review and Assessment Service, a single-payer database capturing >95% of the Korean population. The database included any inpatient encounter with a FS diagnosis from January 2010 to September 2020 for those aged 0-5 years old. We aggregated the monthly number of cases to estimate the incidence per 100,000 patient-years in 2020 (January 1 to September 30) for the same periods in 2010-2019. Results: The incidence of FS in 2020 ranged from 113 per 100,000 (95% confidence interval [CI], 108-118 per 100,000) in January to 27 per 100,000 (95% CI, 25-30 per 100,000) in September, whereas the average FS incidence in 2010-2019 ranged from 116 per 100,000 (95% CI, 112-121 per 100,000) in January to 101 per 100,000 (95% CI, 97-106 per 100,000) in September. Conclusions: The incidence of FS decreased by -38% in 2020, suggesting that social distancing contributed towards decreasing the incidence of FS.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.8 no.5
• /
• pp.303-314
• /
• 2018

In the past few years, government agencies and corporations have succumbed to stealthy, tailored cyberattacks designed to exploit vulnerabilities, disrupt operations and steal valuable information. Security Information and Event Management (SIEM) is useful tool for cyberattacks. SIEM solutions are available in the market but they are too expensive and difficult to use. Then we implemented basic SIEM functions to research and development for future security solutions. We focus on collection, aggregation and analysis of real-time logs from host. This tool allows parsing and search of log data for forensics. Beyond just log management it uses intrusion detection and prioritize of security events inform and support alerting to user. We select Elastic Stack to process and visualization of these security informations. Elastic Stack is a very useful tool for finding information from large data, identifying correlations and creating rich visualizations for monitoring. We suggested using vulnerability check results on our SIEM. We have attacked to the host and got real time user activity for monitoring, alerting and security auditing based this security information management.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.6D
• /
• pp.1003-1010
• /
• 2006

This study developed a traffic accident information management system based on WebGIS that can process a lot of data for giving effectively diagnosis of traffic accidents in serious damage circumstances by traffic accident. Also, this study presents a way to compose and to convey traffic accident information. In addition, non-spatial attributes as well as spatial attributes about traffic accidents information be integrated and managed by the system. To provide Web service, we developed modules that can supply visually spatial information and traffic accidents data through ASP, Javascript, ArcIMS based on Web and constructed a server. And constructed system include a function that offer the now situation of traffic accident in real time, which supply the statistical data of traffic accident through Web as soon as user entry data in comparison with previous way that preparatory period until traffic accidents data is supplied to peoples had been long. Traffic accidents are analyzed with only nonspatial attribute by simply collecting in the past. However, system constructed by this study offer new function that can grasp visually accident spot circumstance and use detailed content and accurate location data as well as statistical data of traffic accidents. Also, it offer interface that can connect directly with accident charge policeman.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.1-21
• /
• 2020

With the rapid development of artificial intelligence technology, various techniques have been developed to extract meaningful information from unstructured text data which constitutes a large portion of big data. Over the past decades, text mining technologies have been utilized in various industries for practical applications. In the field of business intelligence, it has been employed to discover new market and/or technology opportunities and support rational decision making of business participants. The market information such as market size, market growth rate, and market share is essential for setting companies' business strategies. There has been a continuous demand in various fields for specific product level-market information. However, the information has been generally provided at industry level or broad categories based on classification standards, making it difficult to obtain specific and proper information. In this regard, we propose a new methodology that can estimate the market sizes of product groups at more detailed levels than that of previously offered. We applied Word2Vec algorithm, a neural network based semantic word embedding model, to enable automatic market size estimation from individual companies' product information in a bottom-up manner. The overall process is as follows: First, the data related to product information is collected, refined, and restructured into suitable form for applying Word2Vec model. Next, the preprocessed data is embedded into vector space by Word2Vec and then the product groups are derived by extracting similar products names based on cosine similarity calculation. Finally, the sales data on the extracted products is summated to estimate the market size of the product groups. As an experimental data, text data of product names from Statistics Korea's microdata (345,103 cases) were mapped in multidimensional vector space by Word2Vec training. We performed parameters optimization for training and then applied vector dimension of 300 and window size of 15 as optimized parameters for further experiments. We employed index words of Korean Standard Industry Classification (KSIC) as a product name dataset to more efficiently cluster product groups. The product names which are similar to KSIC indexes were extracted based on cosine similarity. The market size of extracted products as one product category was calculated from individual companies' sales data. The market sizes of 11,654 specific product lines were automatically estimated by the proposed model. For the performance verification, the results were compared with actual market size of some items. The Pearson's correlation coefficient was 0.513. Our approach has several advantages differing from the previous studies. First, text mining and machine learning techniques were applied for the first time on market size estimation, overcoming the limitations of traditional sampling based- or multiple assumption required-methods. In addition, the level of market category can be easily and efficiently adjusted according to the purpose of information use by changing cosine similarity threshold. Furthermore, it has a high potential of practical applications since it can resolve unmet needs for detailed market size information in public and private sectors. Specifically, it can be utilized in technology evaluation and technology commercialization support program conducted by governmental institutions, as well as business strategies consulting and market analysis report publishing by private firms. The limitation of our study is that the presented model needs to be improved in terms of accuracy and reliability. The semantic-based word embedding module can be advanced by giving a proper order in the preprocessed dataset or by combining another algorithm such as Jaccard similarity with Word2Vec. Also, the methods of product group clustering can be changed to other types of unsupervised machine learning algorithm. Our group is currently working on subsequent studies and we expect that it can further improve the performance of the conceptually proposed basic model in this study.

• The KIPS Transactions:PartD
• /
• v.15D no.3
• /
• pp.321-326
• /
• 2008

The software defects that are not found in the course of a project frequently appear during the conduct of the maintenance procedure after the complete development of the software. As the frequency of surfacing of defects during the maintenance procedure increases, the cost likewise increases, and the quality and customer reliability decreases. The defect rate will go down only if cause analysis and process improvement are constantly performed. This study embodies the defect tracking system (DTS) by considering the Pareto principle: that most defects are repetitions of defects that have previously occurred. Based on the records of previously occurring defects found during the conduct of a maintenance procedure, DTS tracks the causes of the software defects and provides the developer, operator, and maintenance engineer with the basic data for the improvement of the software concerned so that the defect will no longer be manifested or repeated. The basic function of DTS is to analyze the defect type, provide the measurement index for it, and aggregate the program defect type. Doing these will pave the way for the full correction of all the defects of a software as it will enable the defect correction team to check the measured defect type. When DTS was applied in the software configuration management system of the W company, around 65% of all its software defects were corrected.

• Information and Communications Magazine
• /
• v.11 no.10
• /
• pp.120-138
• /
• 1994

최근에 정보 고속도로 사업계획과 관련하여 세계 여러나라들이 실시하거나 실시 예정인 VOD 시험서비스에 많은 관심이 일고 있다. VOD서비스와 같은 대화형 TV 서비스(I-TV, Interactive-TV Service)는 컴퓨터, 통신 가전업체 그리고 영화 제작사등이 함께 참여하고 있다. 본 기고에서는 차세대 통합 멀티미디어 사업이라는 대화형 TV서비스의 개요와 요소기술, 그리고 서비스의 동향 및 전망을 알아보기로 한다. 1. 대화형 TV 서비스(Interactive TV Service)의 개요 현재까지 대부분의 TV 시청자들은 방송국으로부터 송신되는 공중파나 CATV서비스를 통해 프로그램 서비스를 받아보는 방송(broadcasting)방식 형태의 서비스를 이용해 왔다. 그러나 이러한 일방적인 수신방식은 최근 미국의 클린턴 행정부가 미국의 차세대 기반 통신망 구축 정책(정보 고속도로 사업; Information Super High-way)의 일환으로, CATV 사업자뿐만 아니라 지역 전화회사까지도 참여하고 있는 VOD(Video On Demand) 시험서비스가 시행됨에 따라서 미국은 물론, 우리나라를 비롯한 여러나라에서 대화형식의 TV서비스를 시행하려는 움직임이 일어나고 있다. 이러한 움직임은 지난 ‘92년 7월 16일 미 연방통신위원회(FCC. Federal Communication Committee)가 그동안 전화회사에 대해 CATV 시업진출을 금지하였던 정책을 바꿔서, CATV에 대항한 경쟁도입과 기반 정보통신망 정비를 목표로 하여 비디오 다이얼톤(VDT(주), Video Dial Tone)이라는 이름하에 공중통신 사업자에게도 영상신호를 전송을 인가함에 따라 시작되었다. VOD서비스와 같은 대화형 TV서비스는 가입자가 원하는 서비스를 시간에 관계없이 댁내의 통신회선(일반전화망, 케이블망, 광화이버망, 위성망)을 통해서 요구 즉시 실시간에 서비스를 받아볼 수 있는 멀키미디어의 통합서비스 방식이다. 도한 이 서비스는 음성통신과 방송의 통합형 서비스로서, 방송국, 전화회사 혹은 서비스 제공업자(ISP ; Information Service Provider)에 의해 제공된 디지탈 영상 및 일반 데이터 등을 압축하여 서버에 저장한 후, 소비자의 요청이 있을 경우에 통신회선을 통해 즉시 서비스 해준다. 이와 같은 대화형 TV 서비스의 구성요소로는 다양한 영상정보 및 데이타를 보유하고 있는 정보제공자, 전화선이나 CATV 또는 위성방송과 연결해 주는 지역별 비디오 서버 및 교환장치, 통신회선을 통해 전송된 서비스를 영상정보로 만들어주고 또한 가입자의 요구를 즉시 정보제공자에게 알려줄 수 있는 가정용 단말기 (STB. Set Top Box)로 이루어진다. 여기서, 비디오 서버는 다양한 대화형 TV 서비스를 제공해 주는 중계자로서, 영화, 홈쇼핑, 홈뱅킹, 대화형 교육, 비디오 게임 등을 가정에서도 가능하게 해 줄 뿐 아니라, 가입자를 관리하는 기능을 가지고 있어서 가입자 정보는 물론, 각종 서비스 사용료, 개인 통계자료 분석 등도 집계할 수 있는 기능을 가지고 있다.(그림1) 본 기고에서는 이러한 대화형 TV서비스 중에 대표적 응용서비스인 VOD 서비스에 사용되는 기술요소와 각국에서 진행되고 있는 VOD 서비스 동향 및 전망에 대해 알아보고자 한다. (주) VDT(Video Dial Tone) : FCC가 ‘92년 7월 16일에 인가한 지역 전화회사에 의한 가정용 영상 전송서비스 및 CATV에 대항한 경쟁도입과 Infrastructure의 정비를 목표로 하여 결정한 내용은 다음과 같다. 지역 전화회사에 대하여 (1) 공중통신사업자에 대하여 서비스 제공자에 대한 영상신호의 전송을 인가(기본서비스), (2) Video Gateway 서비스, 비디오 기기 제공, 서비스 제공사업자에 대한 과금, 요금징수 대행을 인가(고도서비스), (3) 프로그램 제공자에 대하여 자본출자율을 5%로 높이는 외에 (종래는 1%) 업무 관계의 확대(합병회사 설립 및 consultant 계약 등)를 인가. (4) Rural 지역(영업지역의회의 지방)에 대한 직접 프로그램 제공의 특별인가(주민이 1만세데 미만의 지방 공공 단체만 가능, 영업지역내에서는 제한없음), (5) 지방공공 사업체에 의한 영업면허의 불요(지역 전화회사가 직접 사용자에 서비스를 제공하지 않기 때문에 CATV 서비스로는 보지 않기 때문). (6)의 회로의 권고(케이블 정책법으로 결정되어 있는 통신사업자와 CATV회사 자본의 상호보유 금지의 해제) 등이다.