• 한국음향학회지
• /
• 제15권2호
• /
• pp.61-71
• /
• 1996

본 논문에서는 음성합성을 위한 대화체(회화체) 음성의 triphone 단위 데이터베이스 구축에 대하여 보고한다. 특히 본 연구는 방송 매체를 이용하여 대화체 음성을 수집하고 3차에 걸친 대화체 표기(transcription)작업을 거쳐 triphone 단위의 분할 및 음성기호층 단계의 레이블링을 목표로 진행되었다. 수집된 총 10시간 방송분량중 6시간 분량을 데이터베이스 구축에 사용하였으며, 나머지 4시간은 예비 분으로 수집되었다. 낭독체 음성 데이터베이스 구축과는 여러 면에서 다른, 대화체 음성 데이터베이스 구축을 위한 음성 데이터 수집에서부터 triphone 단위 레이블링까지의 과정을 본 논문에서 기술하고, 보다 체계적이고 일관성있는 대화체 음성 데이터베이스 구축을 위해 필요한 계획 및 요구 사항에 대해서 논하고자 한다.

• 한국건축시공학회지
• /
• 제3권4호
• /
• pp.95-102
• /
• 2003

A construction project's period is limited by the order manufacture. And a construction project is requested a specific place, and has the characteristic to be irregular. Today, construction is to be representative high-cost and low-efficiency. This kind of problems appears in small construction projects more often than large construction projects. According to this facts, we recognized necessity of making practical application in methods of construction management. Also, we suggest application and establishment of integrated responsibility management department, PMIS(Project Management Information System) and Data Base. From now on, methods of construction management could be good opportunities for small construction projects and local construction companies. Therefore, small and local construction companies should try to improve their abilities. Likewise, the countries systematize to constitute the relation laws.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1989년도 한국자동제어학술회의논문집; Seoul, Korea; 27-28 Oct. 1989
• /
• pp.407-412
• /
• 1989

Using the knowledge base which contains the patterns and data of the past experience of a plant, a learning control method is suggested. The knowledge for controlling a plant is stored to the knowledge base and continually modified after performance evaluation of an applied control input. The performance of the resultant knowledge based control system is examined by an application to process.

• 국제학술발표논문집
• /
• The 7th International Conference on Construction Engineering and Project Management Summit Forum on Sustainable Construction and Management
• /
• pp.291-296
• /
• 2017

Infrastructure development as national project suffers from social conflict which is one of main risk to be managed. Social conflicts have a negative impact on not only the social integration but also the national economy as they require enormous social costs to be solved. Against this backdrop, this study analyzes social conflict using articles published by online news media based on web-crawling and natural language processing (NLP) techniques. As an illustrative case, the Jeju Naval Base (JNB) project which is one of representative conflict case in South Korea is analyzed. Total of 21,788 articles and representative keywords are identified annually. Additionally, comparative analysis is conducted between the extracted keywords and actual events occurred during the project. The authors explain actual events in the JNB project based on the extracted words by the year. This study contributes to analyze social conflict and to extract meaningful information from unstructured data.

• 국제학술발표논문집
• /
• The 2th International Conference on Construction Engineering and Project Management
• /
• pp.433-439
• /
• 2007

A ubiquitous environment in construction requires integrating hardware and software systems. Currently the Construction System Integration Lab (CSIL) at Pusan National University is currently studying the application of ubiquitous technology for better communication in the construction process. In this paper, a pilot of Ubiquitous Concrete Pour System (u-CPS) has been presented to demonstrate the effectiveness of data exchange in the concrete pour process. The u-CPS environment takes advantage of the RFID technology for collecting construction data. The pilot can automatically generate the data for concrete pour work such as departure time, arrival time, concrete pour time. Construction managers can keep track of the progress of concrete pour work using the information. A case study was done for a building construction using the pilot system, the result of which demonstrated that the RFID-base system can help improve the effectiveness of data communication during the concrete pour process.

• 한국지반환경공학회 논문집
• /
• 제14권11호
• /
• pp.47-54
• /
• 2013

본 연구는 모듈러 도로시스템에 적용 가능한 맞춤형 지지구조 모듈러 개발을 위한 기초연구로서 연약지반 조건에 대한 지지구조 모듈의 설계사례 연구를 수행하였다. 지지구조 모듈은 크로스빔 지지구조와 크로스빔-말뚝 지지구조 두 가지 형식으로 제안하였다. 두 형식에 대한 설계검토를 통하여 연약지반에 대해서는 크로스빔-말뚝 지지구조가 적합한 것을 확인하였으며, 모듈러 도로시스템에 적합한 크로스빔-말뚝 지지구조의 제원을 도출하였다. 맞춤형 지지구조 모듈러 개발을 위해서는 향후 다양한 국내 도로건설 현장의 원지반이나 노체에 대한 DB를 구축하고 설계사례 연구를 통하여 지반조건에 따른 지지구조 모듈의 분류 및 체계화가 필요할 것으로 판단된다.

• 건설안전기술
• /
• 통권36호
• /
• pp.73-81
• /
• 2006

Construction fall accidents have been investigated by many researchers. Construction workers are prone to fall when elevations of the construction site is high. And falls are the most fatal accidents: it can be directly linked to the death. Construction fall accidents might be reduced by predetermining several areas which are highly probable to have fall accidents and by controlling such areas until the completion of the building construction. In this paper, a fall prevention system is suggested which can identify the areas where the focus on fall protection is perhaps most needed from the process characteristics. Main methodologies for this research are summarized as follows: 1. A data base on elements and types of falls is constructed from the data analysis of last 10 years fall accidents history. 2. Guideline is derived by identifying the highly probable areas of fall accidents with respect to the specific construction process. 3. Developed system is verified by applying the system to construction sites. 4. Finally a fall prevention system is suggested by utilizing the fall accidents data.

• 한국경영과학회:학술대회논문집
• /
• 대한산업공학회/한국경영과학회 2004년도 춘계공동학술대회 논문집
• /
• pp.9-12
• /
• 2004

Developing an efficient knowledge base construction mechanism as an input method for expert systems (ES) development is of extreme importance due to the fact that an input process takes a lot of time and cost in constructing an ES. Most ES require experts to explicit their tacit knowledge into a form of explicit knowledge base with a full sentence. In addition, the explicit knowledge bases were composed of strict grammar and keywords. To overcome these limitations, this paper proposes a knowledge conceptualization and construction mechanism for automated knowledge acquisition, allowing an efficient decision. To this purpose, we extended traditional knowledge map (KM) construction process to dynamic knowledge map (DKM) and combined this algorithm with relational database (RDB). In the experiment section, we used medical data to show the efficiency of our proposed mechanism. Each rule in the DKM was characterized by the name of disease, clinical attributes and their treatments. Experimental results with various disease show that the proposed system is superior in terms of understanding and convenience of use.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 ITC-CSCC -3
• /
• pp.1764-1767
• /
• 2002

Recent explosive increases in information 'volume have led to a rapid development or a change of information technology which stores, searches, and manages a vast amount of information. It is considered that an effective share and utilization of a large amount of digital information produced by work performances is a pivotal element which can make decisive contributions to a great success of business management. This common property of information reflects a changing social paradigm including a change of business processes. This paper is aimed at designing and embodying the construction of knowledge base in an efficient project management system using unsupervised data mining techniques in order to extract information and utilize it as knowledge about standard data (statistical data, template etc.,), size prediction and a danger precaution notice which are needed for a plan and a scheduling of a new project from data coming from already-established projects.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권3호
• /
• pp.814-836
• /
• 2021

Wildlife ecologists and biologists recapture deer to collect tracking data from deer collars or wait for a drop-off of a deer collar construction that is automatically detached and disconnected. The research teams need to manage a base camp with medical trailers, helicopters, and airplanes to capture deer or wait for several months until the deer collar drops off of the deer's neck. We propose an intelligent robust base-station research with a low-cost and time saving method to obtain recording sensor data from their collars to a listener node, and readings are obtained without opening the weatherproof deer collar. We successfully designed the and implemented a robust base station system for automatically collecting data of the collars and listener motes in harsh wilderness environments. Intelligent solutions were also analyzed for improved data collections and pattern predictions with drone-based detection and tracking algorithms.