• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.461-466
• /
• 2021

본 연구에서는 DECO(Dictionnaire Electronique du COreen) 한국어 전자사전과 LGG(Local-Grammar Graph)에 기반한 반자동 언어데이터 증강(Semi-automatic Symbolic Propagation: SSP) 방식에 입각하여, 핀테크 분야의 CS(Customer Service) 챗봇 NLU(Natural Language Understanding)을 위한 주석 학습 데이터를 효과적으로 생성하고, 이를 기반으로 RASA 오픈 소스에서 제공하는 DIET(Dual Intent and Entity Transformer) 아키텍처를 활용하여 핀테크 CS 챗봇 NLU 시스템을 구현하였다. 실 데이터을 통해 확인된 핀테크 분야의 32가지의 토픽 유형 및 38가지의 핵심 이벤트와 10가지 담화소 구성에 따라, DECO-LGG 데이터 생성 모듈은 질의 및 불만 화행에 대한 양질의 주석 학습 데이터를 효과적으로 생성하며, 이를 의도 분류 및 Slot-filling을 위한 개체명 인식을 종합적으로 처리하는 End to End 방식의 멀티태스크 트랜스포머 모델 DIET로 학습함으로써 DIET-only F1-score 0.931(Intent)/0.865(Slot/Entity), DIET+KoBERT F1-score 0.951(Intent)/0.901(Slot/Entity)의 성능을 확인하였으며, DECO-LGG 기반의 SSP 생성 데이터의 학습 데이터로서의 효과성과 함께 KoBERT에 기반한 DIET 모델 성능의 우수성을 입증하였다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.10
• /
• pp.1905-1913
• /
• 2006

The researches about the hardware and the software implementing ubiquitous sensor network have great rush in recent years. This paper deals the development of a sensor node with the dual interface which also has an RF wireless interface while has an Bluetooth interface used widely in present. This sensor node includes a Atmeg32 microcontroller, a Bluetooth module, a RF module. a temperature-humidity sensor. and I also develop the F/Ws controlling each modules with C language using GCC compiler. The sensor node developed can reaches 15m with Bluetooth interface and 60m with RF interface. It works stably with the voltage above 5V and it consumes currents 21mA average in idle mode, 63mA average in active mode.

• Journal of Power Electronics
• /
• v.16 no.2
• /
• pp.553-563
• /
• 2016

This paper presents a digital hardware implementation of a real-time simulator for a multiphase drive using a field-programmable gate array (FPGA) device. The simulator was developed with a modular and hierarchical design using very high-speed integrated circuit hardware description language (VHDL). Hence, this simulator is flexible and portable. A state-space representation model suitable for FPGA implementations was proposed for a dual three-phase induction machine (DTPIM). The simulator also models a two-level 12-pulse insulated-gate bipolar transistor (IGBT)-based voltage-source converter (VSC), a pulse-width modulation scheme, and a measurement system. Real-time simulation outputs (stator currents and rotor speed) were validated under steady-state and transient conditions using as reference an experimental test bench based on a DTPIM with 15 kW-rated power. The accuracy of the proposed digital hardware implementation was evaluated according to the simulation and experimental results. Finally, statistical performance parameters were provided to analyze the efficiency of the proposed DTPIM hardware implementation method.

• Annual Conference on Human and Language Technology
• /
• 2014.10a
• /
• pp.95-99
• /
• 2014

본 논문에서는 최근 급속히 증가하여 사회적 이슈가 되고 있는 SMS 스팸 필터링을 위한 듀얼 SMS 스팸필터링 기법을 제안한다. 지속적으로 증가하고 새롭게 변형되는 SMS 문자 필터링을 위해서는 패턴 및 스팸 단어 사전을 통한 필터링은 많은 수작업을 요구하여 부적합하다. 그리하여 기계 학습을 이용한 자동화 시스템 구축이 요구되고 있으며, 효과적인 기계 학습을 위해서는 자질 선택과 자질의 가중치 책정 방법이 중요하다. 하지만 SMS 문자 특성상 문장들이 짧기 때문에 출현하는 자질의 수가 적어 분류의 어려움을 겪게 된다. 이 같은 문제를 개선하기 위하여 본 논문에서는 슬라이딩 윈도우 기반 N-gram 확장을 통해 자질을 확장하고, 확장된 자질로 그래프를 구축하여 얕은 구조적 특징을 표현한다. 학습 데이터에 출현한 N-gram 자질을 정점(Vertex)으로, 자질의 출현 빈도를 그래프의 간선(Edge)의 가중치로 설정하여 햄(HAM)과 스팸(SPAM) 그래프를 각각 구성한다. 이렇게 구성된 그래프를 바탕으로 노드의 중요도와 간선의 가중치를 활용하여 최종적인 자질의 가중치를 결정한다. 입력 문자가 도착하면 스팸과 햄의 그래프를 각각 이용하여 입력 문자의 2개의 자질 벡터(Vector)를 생성한다. 생성된 자질 벡터를 지지 벡터 기계(Support Vector Machine)를 이용하여 각 SVM 확률 값(Probability Score)을 얻어 스팸 여부를 결정한다. 3가지의 실험환경에서 바이그램 자질과 이진 가중치를 사용한 기본 시스템보다 F1-Score의 약 최대 2.7%, 최소 0.5%까지 향상되었으며, 결과적으로 평균 약 1.35%의 성능 향상을 얻을 수 있었다.

• Annual Conference on Human and Language Technology
• /
• 1999.10e
• /
• pp.351-358
• /
• 1999

Two experiments were conducted to investigate effects of word-frequency on semantic processing of Hangul. Stimuli were two syllable words, and exemplars and target words were different in the final consonant of the second syllable in the Exp 1 and in the final consonant of the first syllable in the Exp2. Exp 1 shows the results that subjects made more errors on low frequency target words and took longer times on high frequency exemplars than on controls. In Exp 2 subjects took longer times on high frequency examplar-low frequency target word conditions than on controls. These results support the predictions of dual process models and suggest that the use of phonological and visual information depends on word frequency. Phonological activation appears to be an optional rather than obligatory process.

• Korean Journal of Childcare and Education
• /
• v.13 no.2
• /
• pp.149-164
• /
• 2017

Objective: This study aimed to examine the mediation effects of parenting style and daily activity on the relationships between cumulative parental stress and cognitive and language development of children. Methods: The participants of this study were 1,154 mothers, their five-year-old children, and teachers enrolled in the population-based study of the Panel Study of Korea Children (PSKC) conducted by the Korea Institute of Childcare and Education from 2008 to 2013. In order to analyze the data, descriptive analysis, Pearson correlation analysis, and structural equation modeling were used. Results: The results revealed that warm parenting style as well as daily activity had dual mediating effects between cumulative parental stress and child development (cognitive and language development). However, warm parenting style did not have any direct effects on child development in this model. Conclusion/Implications: Given that actual mother-child daily activity had an important role in affecting child development, support programs for parenting should include activities that include specific materials and activities which parents can do with their children in everyday life.

• Korean Journal of Child Studies
• /
• v.18 no.2
• /
• pp.57-71
• /
• 1997

This study discusses the nature of data-interpretation process in the perspective of modern hermeneutics and explores its methodological implications for qualitative research on the life of children. Discussions center around the basic concepts of modern hermeneutics, such as the conceptual distinctions between understanding and explanation, the intersubjective reality constituted between the child's horizon and the researcher's, the concept of the hermeneutics circle and the importance of the openness of language in understanding children's lifeworld. The conclusions are made as follows: First, the concept of "understanding", which is the basic concern of qualitative research on children, has a much different meaning from that of "explanation" in quantitative research. The task of understanding lies in interpreting the lived meaning in context as experienced by children in all its complexity. Second the researcher's subjectivity in the interpretation process has a dual function in that it can produce a faulty hermeneutics circle and yet it provides the researcher with the strong momentum to open up her understanding towards a deeper and wider level. Third, the reliability and the validity of qualitative research should be discussed within the interpretation process itself in that interpretation in qualitative research is guided basically by the researcher's internal dialectic: between data (text) and her subjectivity. Lastly, the language in interpretive research should be open and free in terms of style in order to present a fuller and richer description of children's lifeworld and the delicate textures of their lived meanings.

• Industrial Engineering and Management Systems
• /
• v.8 no.2
• /
• pp.121-130
• /
• 2009

Subjective evaluation has always been regarded as a branch of social science research. Hence, in scientific and especially engineering points of view, its development is always taken for granted despite the importance of its effects on the design and development decisions. In the past, at least two automotive seat survey questionnaires have been developed with high statistical validity and reliability. Nonetheless, both were not local while subjective perception very much depends on demographic background factors. It is felt that since vehicle seat comfort is an important aspect in a seat design, a local sense should be put into the survey that is used as the subjective tool. The proposed vehicle seat discomfort survey questionnaire was developed in dual languages; English and Malay. Malay language is the national language of Malaysia, where the survey was tested. Beside inputs from literatures, key informant interviews helped in establishing the appropriate terms used as survey items. Three experimental runs on two different seats by 22 paid subjects showed that the developed questionnaire is reliable and valid. Furthermore, criterion validity analysis on the survey and previously developed survey showed significant correlation at 0.01 significance level.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2001.06a
• /
• pp.237-257
• /
• 2001

This paper investigates the idiosyncratic understanding of universal quantifiers such as every, each or all by young children at the ages of 4 to 7, and argues that the phenomenon is explicable in terms of the maturation of both the cognitive system and the linguistic system. Evidence for this dual explanation comes from the fact that the visual input, a picture, plays a key role in determining the children’s conceptual representation, suggesting the need for the central integration of visual and linguistic elements; and from the fact that a quantifier in the linguistic input has an intrinsic property, i.e. a <+focus> feature. I have tried to explain the nature of the cognitive factors in terms of the function of the central system, suggesting a modified form of Smith & Tsimpli’s (1995) version of Fodor’s (1983) modularity hypothesis. Conceptual representations of two kinds are in competition with each other and they are integrated into a neutral LOT (Language of Thought) representation at some point . In the process of this integration, the representations from the visual input predominate over those from the auditory input, though the quantize. (treated as new information provided by the latter) is salient in the final representations. When visual conceptual representations predominate over purely linguistic ones, quantifier spreading errors occur. By contrast, when the relevant grammatical knowledge has developed sufficiently to counteract the conceptual representations, this peculiar behaviour by children should disappear. It is argued that children have to learn two kinds of grammatical fact with regard to universal quantification: (i) they have to learn the status of the quantifier as a functional head of DP so that it has to be positioned inside DP; and (ii) they have to learn the Left-Branch Condition which specifies that movement of an element in the left-branch position is possible only by pied-piping the entire phrase.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.13 no.4
• /
• pp.229-237
• /
• 2013

As demand for IT services increase, which are based on SOA and cloud computing, service level agreements (SLAs) have received more attention in the parties concerned. An SLA is usually a paper contract written in natural language. SLA management tools which are commercially available, implement SLAs implicitly in the application with a procedural language. This makes automation of SLA management difficult. It is also laborious to maintain contract management systems because changes in a contract give rise to extensive modifications in the source code. We see the source of the trouble is the existence of documentary SLAs (paper contracts) and corresponding executable SLAs (contracts coded in the procedural language). In this paper, to resolve the current SLA management problems we propose an active SLM (Active Service Level Management) system, which is based on the active SLA (Active Service Level Agreement). In the proposed system, the separated management and processing of dual SLAs can be unified into a single process with the introduction of active SLAs (ASLAs).