• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.33 no.3
• /
• pp.347-370
• /
• 2022

Most public library websites provide unstructured cultural program data, which cannot be produced and utilized systematically as bibliographic information. It is not sufficiently used in existing library big data research or cases, and there is a risk of disappearing when the website is reorganized or the person in charge is changed. This study developed a data schema that can be used in conjunction with bibliographic data by collecting and analyzing cultural programs and community service data produced in an unstructured manner and proposed to share and utilize public library cultural programs and community service data, and establish a library big data platform that can serve as an information channel between librarians who are cultural program planners. Library program data posted on the library website can be integrated and managed through the platform, securing continuity of work, and systematically managing and preserving the specialized service history of individual libraries.

• The Korean Journal of Applied Statistics
• /
• v.31 no.5
• /
• pp.655-665
• /
• 2018

Social network services (SNS) have become the most common tool for the communication of public and private opinions as well as public issues; consequently, one may form or drive public opinions to advocate by spreading positive content using SNS. Controversy for survey data based opinion poll accuracy continues in relation to response rate or sampling methodology. This study suggests complementary measures that additionally consider the sentiment analysis results of unstructured data on a social network by data crawling and sentiment dictionary adjustment process. The suggested method shows the improvement of prediction accuracy by decreasing error rates.

• Journal of Information Technology and Architecture
• /
• v.11 no.1
• /
• pp.63-73
• /
• 2014

As the advance of WWW, unstructured data including texts are taking users' interests more and more. These unstructured data created by WWW users represent users' subjective opinions thus we can get very useful information such as users' personal tastes or perspectives from them if we analyze appropriately. In this paper, we provide various analysis efficiently for unstructured text documents by taking advantage of OLAP (On-Line Analytical Processing) multidimensional cube technology. OLAP cubes have been widely used for the multidimensional analysis for structured data such as simple alphabetic and numberic data but they didn't have used for unstructured data consisting of long texts. In order to provide multidimensional analysis for unstructured text data, however, Text Cube model has been proposed precently. It incorporates term frequency and inverted index as measurements to search and analyze text databases which play key roles in information retrieval. The primary goal of this paper is to apply this text cube model to a real data set from in an Internet site sharing hotel information and to provide multidimensional analysis for users' reviews on hotels written in texts. To achieve this goal, we first build text cubes for the hotel review data. By using the text cubes, we design and implement the system which provides multidimensional keyword search features to search and to analyze review texts on various dimensions. This system will be able to help users to get valuable guest-subjective summary information easily. Furthermore, this paper evaluats the proposed systems through various experiments and it reveals the effectiveness of the system.

• Water for future
• /
• v.56 no.8
• /
• pp.31-36
• /
• 2023

• Korean Journal of Construction Engineering and Management
• /
• v.14 no.5
• /
• pp.35-43
• /
• 2013

Nowadays the widely used media in architecture include visualizations, animations and three-dimensional models. 3D digital methods using active CAM(Computer Aided Manufacturing) and CNC(Computerized Numerical Control) imaging have been developed for accurate shape and 3D measurements in freeform buildings. In contrast to a conventional building using auto CAD system and others, the proposed digital optimization method is based on a combination of 3D numerical data and parametric 3D model for design and construction. The objective of this paper is therefore to present digital optimization process for constructability of freeform building. The method can be useful in the effective implementation of an error-proofing process of freeform building during design and construction phase. 3D digital coordinate data can be used effectively to identify correct size of structural and finish members and installation location of each members in construction field. In addition, architects, engineers and contractors can evaluate design, materials, constructability and identify error-proofing opportunities. Other project participants can also include representatives from all levels of management, departments as well as workers and key subcontractors' personnel, if necessary. The 3D digital optimization process is therefore appropriate to serious variations in freeform shape. For future study, the developed digital optimization method is necessary to be carried out to verify the robustness and accuracy for constructability in construction field.

• The Journal of Society for e-Business Studies
• /
• v.21 no.2
• /
• pp.151-161
• /
• 2016

In the era of the Web 2.0, characterized by the openness, sharing and participation, it is easy for internet users to produce and share the data. The amount of the unstructured data which occupies most of the digital world's data has increased exponentially. One of the kinds of the unstructured data called personal online product reviews is necessary for both the company that produces those products and the potential customers who are interested in those products. In order to extract useful information from lots of scattered review data, the process of collecting data, storing, preprocessing, analyzing, and drawing a conclusion is needed. Therefore we introduce the text-mining methodology for applying the natural language process technology to the text format data like product review in order to carry out extracting structured data by using R programming. Also, we introduce the data-mining to derive the purpose-specific customized information from the structured review information drawn by the text-mining.

• Korean Journal of Construction Engineering and Management
• /
• v.12 no.3
• /
• pp.11-21
• /
• 2011

The most critical issue in free-form buildings is how to construct the free-formed exterior facade panels. Their geometric complexity delivers many cons and problems in fabricating and constructing their shapes. To construct a free-form building, first of all, its skin has to be chopped into small pieces, which is called panelization. After panelization, the panels go through an optimization process to construct them economically. The panel's geometries are modified or regenerated through this optimization process. In this study, the panel optimization process of free-form buildings are performed through a case study. The panel shapes of the case study are modeled with Digital Project. To test the constructability of the various panels, 8 mock-up panels are made and laser scanning technology is applied to measure the preciseness of the panels manufactured in comparison with their original design.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.501-503
• /
• 2019

IT 기술의 발달에 따라 다양한 웹 미디어의 데이터가 기하급수적으로 증가하고 있으며 이는 비정형 형태의 빅 데이터로 활용도가 매우 높다. 그 중 인터넷 뉴스나 SNS 등은 시간의 흐름에 따라 다양한 이슈들이 서로 영향을 주며 발생, 결합, 분화, 소멸된다. 본 논문에서는 인터넷상에서 발생하는 비정형 데이터들을 수집하여 텍스트 마이닝을 통해 글의 주요이슈 키워드, 카테고리, 날짜 등을 추출한다. 추출한 데이터를 일정 기간별로 나누어 이슈 매핑을 통해 이슈간의 상관관계를 분석한다. 나아가 LSTM 또는 GRU를 이용한 딥러닝을 통해 앞으로의 이슈를 예측하는 시스템 설계를 제안한다.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.309-309
• /
• 2018

기후변화에 따른 강우의 규모와 발생빈도 증가로 농촌유역의 홍수 피해는 지속적으로 증가하고 있다. 하지만 우리나라의 홍수 피해 저감 대책은 도시지역의 대하천 주변으로 집중되어있으며, 소하천 및 농촌유역의 홍수 피해 저감에 대한 관리와 투자 노력은 부족한 실정이다. 특히, 최근 들어 갑작스런 집중호우 등으로 인한 농촌유역 돌발홍수 피해 사례가 증가하고 있으며, 이에 대응하기 위해서는 홍수 발생 등을 신속하게 파악하기 위한 돌발홍수 예경보 시스템 개발이 필요하다. 한편, 최근 산업의 혁신과 생산성 향상을 위한 새로운 패러다임으로 4차 산업혁명이 대두되고 있으며, 빅데이터와 인공지능 (Artificial Intelligence, AI)을 비롯하여 사물인터넷 (Internet of Things, IoT), 드론, 슈퍼컴퓨팅 등의 이른바 4차 산업혁명 기술을 활용한 연구가 수행되고 있다. 본 연구에서는 기후변화에 따른 농촌유역 홍수 피해를 저감하고 또한 사전에 대비하기 위해 빅데이터와 인공지능 등 4차 산업혁명 기술을 적용한 농촌유역 돌발홍수 예경보 시스템을 개발하고 그 적용성을 평가하고자 한다. 우선, 농촌유역의 홍수와 관련된 빅데이터 (기상 자료, 수문 자료, 기후변화 자료, 농업용 수리구조물 자료 등)를 토대로 정형 빅데이터와 비정형 빅데이터를 구분 추출하고 이를 연계 해석할 수 있는 시스템을 개발하였다. 추출한 정형 및 비정형 빅데이터를 활용하여 딥러닝을 기반으로 농촌유역의 홍수를 예측하고 홍수 예경보 기준에 따른 평가를 수행할 수 있는 시스템을 개발하였다. 과거 강우사상을 홍수 예경보 시스템에 적용하여 홍수 모의 결과를 도출하였으며, 재해연보 등과 비교 분석하여 시스템의 적용성을 분석하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2015.07a
• /
• pp.301-303
• /
• 2015

국내 건설행정 투명화 및 경쟁력 향상 목적으로 개발된 건설사업정보시스템에 정부와 운영주체는 다양한 기능개선 방안과 관련 연구를 수행하며 시스템 성능을 개선시켜왔다. 그러나 기 추진된 성능향상 방안이 공공업무 처리에 중점 되어 대국민 사용자를 위한 콘텐츠 및 기능 등의 서비스가 미흡한 상황이다. 이에 본 논문에서는 건설사업정보 건설인허가시스템의 도로점용장소별 허가현황 기능을 중심으로 빅 데이터를 이용한 허가현황 정보 제공 방안을 제안하였다. 제안한 기능개선 방안은 기 구축된 비정형 데이터를 빅 데이터 기반으로 재분석하여 구글 맵에 가시화함으로써 공공업무 데이터 처리 뿐만 아니라 대국민 서비스를 위한 콘텐츠 제공이 가능하도록 하였다. 뿐만 아니라 그동안 축적된 15TB이상의 건설관련 데이터의 재활용 가능성을 시사함으로써 시스템 활용성 증대 및 개편 방향에 도움이 될 것으로 판단된다.