• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.353-360
• /
• 2021

음성 대화 시스템에서는 사용자가 잘못된 슬롯명을 말하거나 음성인식 오류가 발생해 사용자의 의도에 맞지 않는 응답을 하는 경우가 있다. 이러한 문제를 해결하고자 말뭉치나 사전 데이터를 활용한 질의 교정 방법들이 제안되지만, 이는 지속적으로 사람이 개입하여 데이터를 주입해야하는 한계가 있다. 본 논문에서는 축적된 로그 데이터를 활용하여 사람의 개입 없이 음악 재생에 필요한 슬롯을 교정하는 자기 학습(Self-learning) 기반의 모델을 제안한다. 이 모델은 사용자가 특정 음악을 재생하고자 유사한 질의를 반복하는 상황을 이용하여 비지도 학습 기반으로 학습하고 음악 재생에 실패한 슬롯을 교정한다. 그리고, 학습한 모델 결과의 정확도에 대한 불확실성을 해소하기 위해 질의 슬롯 관계 유사도 모델을 이용하여 교정 결과에 대한 검증을 하고 슬롯 교정 결과에 대한 안정성을 보장한다. 모델 학습을 위한 데이터셋은 사용자가 연속으로 질의한 세션 데이터로부터 추출하며, 음악 재생 슬롯 세션 데이터와 질의 슬롯 관계 유사도 데이터를 각각 구축하여 슬롯 교정 모델과 질의 슬롯 관계 유사도 모델을 학습한다. 교정된 슬롯을 분석한 결과 발음 정보가 유사한 슬롯 뿐만 아니라 의미적인 관계가 있는 슬롯으로도 교정하여 사전 기반 방식보다 다양한 유형의 교정이 가능한 것을 보였다. 3 개월 간 수집된 로그 데이터로 학습한 음악 재생 슬롯 교정 모델은 일주일 동안 반복한 고유 질의 기준, 음악 재생 실패의 12%를 개선하는 성능을 보였다.

• Proceedings of the Korean Institute Of Construction Engineering and Management
• /
• 2004.11a
• /
• pp.101-108
• /
• 2004

The results of the study include a computerized system and a systematic Process model for risk management and analysis. This study analyzes the present status of risk management in the construction industry, and then suggests reasonable methods for improved risk management plans. This study defines risk management procedures as preparation, identification, analysis, response and management to manage potential risks In the construction project. The modules for computerizing this system consist of planning, construction, application of WBS (Work Breakdown Structure) and RBS (Risk Breakdown Structure), and risk analysis. The method logy for analyzing construction risk uses fuzzy theory, and the scope of developed system is focused to the contractors. The risk management system suggested in this study operates on the Internet, for providing contractors with a useful risk management tool by online system, with web-based menus that is helpful for practical application.

• Journal of KIISE:Software and Applications
• /
• v.35 no.7
• /
• pp.431-441
• /
• 2008

The academic information analysis service is including automatic recommendation of reviewers and analysis of researchers' accomplishments. The service of recommendation of reviewers should be processed in a transparent, fair and accountable way. When selecting reviewers, the following information must be considered: subject of project, reviewer's maj or, expertness of reviewer, relationship between applicant and reviewer. The analysis service of researchers' accomplishments is providing statistic information of researcher, institution and location based on accomplishments including book, article, patent, report and work of art. In order to support these services, we designed ontology for academic information, converted legacy data to RDF triples, expanded knowledge appropriate to services using OntoFrame. OntoFrame is service framework which includes ontology, reasoning engine, triple store. In our study, we propose the design methodology of ontology and service system for academic information based on OntoFrame. And then we explain the components of service system, processing steps of automatic recommendation of reviewers and analysis of researchers' accomplishments.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.26 no.3D
• /
• pp.469-480
• /
• 2006

The results of the study include a computerized system and a systematic process model for risk management and analysis. This study analyzes the present status of risk management in the construction industry, and then suggests reasonable methods for improved risk management plans. This study defines risk management procedures as preparation, identification, analysis, response and management to manage potential risks in the construction project. The modules for computerizing in this system consist of planning, construction, application of WBS (Work Breakdown Structure) and RBS (Risk Breakdown Structure), and risk analysis. The methodology for analyzing construction risk uses fuzzy theory, and the scope of developed system is focused to the contractors. The risk management system suggested in this study operates on the Internet, for providing contractors with a useful risk management tool by online system, with web-based menus that is helpful for practical application.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.7
• /
• pp.1276-1284
• /
• 2017

Lightweight Encryption Algorithm (LEA) that was standardized as a lightweight block cipher was implemented with 8-bit data path, and the vulnerability of LEA encryption processor to correlation power analysis (CPA) attack was analyzed. The CPA used in this paper detects correct round keys by analyzing correlation coefficient between the Hamming distance of the computed data by applying hypothesized keys and the power dissipated in LEA crypto-processor. As a result of CPA attack, correct round keys were detected, which have maximum correlation coefficients of 0.6937, 0.5507, and this experimental result shows that block cipher LEA is vulnerable to power analysis attacks. A masking method based on TRNG was proposed as a countermeasure to CPA attack. By applying masking method that adds random values obtained from TRNG to the intermediate data of encryption, incorrect round keys having maximum correlation coefficients of 0.1293, 0.1190 were analyzed. It means that the proposed masking method is an effective countermeasure to CPA attack.

• The Journal of the Korea Contents Association
• /
• v.15 no.8
• /
• pp.611-624
• /
• 2015

Humanities-collaborated practices in various fields have been supported by several governmental policies as a way to overcome recent social problems. Museum, due to its roles in preserving and sharing cultural and historical heritage for the development of society and communities, has emerged as a key base for the humanities-collaborated practices. This study, in this context, aimed to analyze how the humanities-based museum education programs have been practiced in the museums, which, in turn, might provide a basic data for the future direction and development of the programs and the related governmental policy as well. For this purpose, this study first examined theoretical framework for the link between humanities and museum education, from which a survey was developed. The survey was conducted to 75 museum educators who have been in charge of humanities-based educational programs. The result showed that most programs seemed to reflect the basic direction and purpose of the humanities-linked with museum education. Yet, more needs were found in terms of the educational themes and methods which can facilitate multicultural literacy and reflective thinking and activities within experiential learning environments.

• Clinical and Experimental Reproductive Medicine
• /
• v.32 no.4
• /
• pp.293-300
• /
• 2005

연구목적: 단일 유전자 이상에 대한 착상전 유전진단을 성공적으로 시행하기 위해서는 효과적이고 신뢰도가 높은 PCR 방법의 확립이 중요하다. 본 연구에서는 alkaline lysis와 duplex nested PCR 방법을 단일 림프구와 할구의 유전자 분석에 적용하여 그 효용성을 확인하고자 하였다. 재료 및 방법: 단일 유전자의 이상이 확인된 Duchenne muscular dystrophy (DMD), ornithine transcarbamylase (OTC) 결핍증과 epidermolysis bullosa (EB) 가계의 대상자들에서 채취한 단일 림프구와 공여 받은 배아의 할구를 이용하여 각각 PCR, restriction fragment length polymorphism (RFLP)와 direct DNA sequencing 분석을 시행하였다. 이러한 분석에서 유전자 증폭률 (amplification rate)과 두개의 allele 중에서 하나의 allele이 증폭되지 않는 allele drop-out (ADO) 빈도에 대해 살펴보았다. 결 과: 단일 림프구와 할구를 이용한 PCR 방법의 유전자 증폭률은 DMD에서 91.1%와 81.8%, OTC 결핍증에서 96.0%와 78.1%, EB에서 91.3%와 90.0%를 각각 나타냈으며, ADO 빈도는 OTC 결핍증에서 13.3%, EB에서 16.8%로 관찰되었다. 결 론: 본 연구에서 적용한 alkaline lysis와 duplex nested PCR 방법은 단일 유전자에 대한 착상전 유전진단에 성공적으로 적용할 수 있는 방법으로 생각되며, ADO 빈도를 최소화할 수 있는 효율적인 방법의 개발에 대한 지속적인 연구가 필요하다.

• 한국신재생에너지학회:학술대회논문집
• /
• 2009.11a
• /
• pp.381-381
• /
• 2009

염료감응형 태양전지의 효율 향상을 위한 다양한 방법들 중 $TiO_2$ 나노 파우더의 표면 개질 및 페이스트의 분산성 향상을 위한 연구가 활발하게 진행되고 있다. 기존 나노 파우더의 표면 개질법으로는 액상 공정인 졸겔법이 있으나 표면 처리 공정에서의 응집현상은 아직 해결해야 할 과제 중 하나이다. 이에 본 연구에서는 진공증착방법인 ALD법을 이용하여 염료감응형 태양전지용 $TiO_2$ 나노 파우더의 $SiO_2$ 산화물 표면처리를 통한 분산특성을 파악하였다. 기존 ALD법의 경우 reactor의 온도가 $300{\sim}500^{\circ}C$ 정도의 고온에서 공정이 이루어졌지만 본 실험에서는 2차 아민계촉매(pyridine)을 사용하여 reactor의 온도를 $30^{\circ}C$정도의 저온공정에서 $SiO_2$ 산화물을 코팅을 하였다. MO source로는 액체상태의 TEOS$(Si(OC_2H_5)_4)$를, 반응가스로는 $H_2O$를 사용하였고, 불활성 기체인 Ar 가스는 purge 가스로 각각 사용 하였다. ALD 공정에 의해 표면처리 된 $TiO_2$ 나노 파우더의 분산특성은 각 공정 cycle에 따라 FESEM을 통하여 입자의 형상 및 분산성을 확인하였으며 입도 분석기를 통하여 부피의 변화 및 분산 특성을 확인하였다. 공정 cycle 이 증가함에 따라 입자간의 응집현상이 개선되는 것을 확인 할 수 있었으며, 100cycles에서 응집현상이 가장 많이 감소하는 것을 확인할 수 있었다. 또한 표면 처리된 $SiO_2$ 산화막은 XRD를 통한 결정 분석 및 EDX를 통한 정성 분석을 통하여 확인하였다.

• Polymer(Korea)
• /
• v.33 no.4
• /
• pp.307-312
• /
• 2009

The fabrication of PHBV nanofibers containing various plant polyphenols by electrospinning has been examined. It has been found that the average diameters of fibers increased by the adding of polyphenols. The resulting fibers exhibited a uniform diameter ranging from 340 to 450 nm. As the concentration of polyphenols increased, the diameter of fibers increased due to the hydrogen bonding interaction between the ester groups of PHBV and hydroxyl groups of polyphenols. The interaction between PHBV and polyphenols, which forms a complex together in solution, was verifed by UV measurement. ATR-FTIR analysis confirmed the existence of the hydrogen bonding interaction. The semicrystalline structure of the PHBV nanofiber was observed from XRD pattern. The crystallinity of PHBV nanofibers was increased by the adding of polyphenols. PHBV nanofibers containing polyphenols showed superior antimicrobial activities.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.5
• /
• pp.183-191
• /
• 2010

Unique and invariant biometric characteristics have been used for secure user authentication. Storing original biometric data is not acceptable due to privacy and security concerns of biometric technology. In order to enhance the security of the biometric data, the cancelable biometrics was introduced. Using revocable and non-invertible transformation, the cancelable biometrics can provide a way of more secure biometric authentication. In this paper, we present a new cancelable bits extraction method for the facial data. For the feature extraction, the Subpattern-based Principle Component Analysis (PCA) is adopted. The Subpattern-based PCA divides a whole image into a set of partitioned subpatterns and extracts principle components from each subpattern area. The feature extracted by using Subpattern-based PCA is discretized with a helper data based method. The elements of the obtained bits are evaluated and ordered according to a measure based on the fisher criterion. Finally, the most discriminative bits are chosen as the biometric bits string and used for authentication of each identity. Even if the generated bits string is compromised, new bits string can be generated simply by changing the helper data. Because, the helper data utilizes partial information of the feature, the proposed method does not reveal privacy sensitive biometric information of the user. For a security evaluation of the proposed method, a scenario in which the helper is compromised by an adversary is also considered.