• The KIPS Transactions:PartB
• /
• v.12B no.3 s.99
• /
• pp.365-368
• /
• 2005

We propose a speech act analysis method for Korean dialogue using Support Vector Machines (SVM). We use a lexical form of a word, its part of speech (POS) tags, and bigrams of POS tags as sentence features and the contexts of the previous utterance as context features. We select informative features by Chi square statistics. After training SVM with the selected features, SVM classifiers determine the speech act of each utterance. In experiment, we acquired overall $90.54\%$ of accuracy with dialogue corpus for hotel reservation domain.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.8
• /
• pp.2113-2123
• /
• 1999

The partial free order feature of Korean makes grammar size represented by CFG too big and that's why grammar has to contain all the ordered words. There are some problems to parse spoken language, because spontaneous spoken language has special features such as meaningless words, repetitious speech, etc. So, in this paper, we define 'Core-Concept' as the necessary element for parsing and we describe grammar only using Core-Concept. And we can prevent grammar from becoming very large and reduce an additional parsing burden as we select. Core-Concept described in grammar as parsing element. Through this strategy, we present that the simplified grammar can give us more efficient method to get right results. Experiments show that our parsing strategy has an average of 98% or over success rate in correct parsing results.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.11b
• /
• pp.733-735
• /
• 2005

최근 음성인식 분야 에서는 잡음 환경에서 좀 더 신뢰도 높은 음성 인식 결과물 얻기 위하여 인식 결과 도출 단계에서 여러 가지 정보를 융합 하는 방법이나 인식결과를 후처리 하여 새로운 결과를 얻어 내는 방법들이 연구 되고 있다. 본 논문에서는 개인 모바일 기기에서의 음성 인식 환경에서 사용자의 발화 패턴 정보를 가지는 문맥 정보를 활용함으로서 잡음 환경에서의 음성 정보 손실에 따른 인식률 하락을 보완하는 방법을 제안한다. 먼저 사용자의 기기 사용 로그나 발화 로그 정보로부터 특정 명령어들의 순차적 발화 패턴을 마이닝하여 문맥 정보를 구성한다. 이 후 음성 발화시에 인식기의 최종 인식 결과에 대한 신뢰도가 떨어진다고 판단될 때 앞서 얻어진 문맥 정보의 신뢰도를 인식기의 각 후보단어들의 인식률과 융합하여 새로운 인식 결과를 도출해 낸다. 이러한 과정에서 인식기 결과에 대한 신뢰성을 판단하는 기준을 실험을 통하여 결정 하였으며 신뢰성이 기준 이하일 경우의 융합 과정을 위하여 후보 단어 인식률과 문맥정보를 적절히 융합할 수 있는 방법을 제안한다.

• Annual Conference on Human and Language Technology
• /
• 2011.10a
• /
• pp.133-135
• /
• 2011

본 논문은 다중 도메인 대화 시스템에서 사용자의 발화에 가장 적합한 도메인을 결정하는 방법에 관하여 논한다. 다중 도메인 대화 시스템 구축 시, 도메인의 확장성 재고 및 각 도메인 별 특성의 효과적인 적용을 위해, 각 도메인 별 단일 도메인 대화 전문가를 구축하고, 다수의 도메인 대화 전문가들을 통합함으로써, 다양한 도메인을 처리할 수 있는 대화 시스템을 구축할 수 있다. 이 때, 자연스러운 대화 처리를 위해서 주어진 사용자의 발화에 가장 적합한 도메인을 결정하는 기술이 필요하다. 본 논문에서는 사용자 발화의 의도 분석 결과 및 이전 사용자 발화 도메인 정보를 이용하는 구축된 분류 모델에 기반한 도메인 결정 방법을 제안하고, 실험 결과를 통해 제안된 방법의 효과를 검증한다.

• Journal of the Korean Institute of Gas
• /
• v.16 no.1
• /
• pp.26-32
• /
• 2012

The autoignition temperature is one of the most important physical properties used to determine the flammability characteristics of chemical substances. Despite the needs of the experimental autoignition temperature data for the design of chemical plants, it is not easy to get the data. This study have built and compared partial least squares (PLS) and support vector machine (SVM) models to predict the autoignition temperatures of 503 organic compounds out of DIPPR 801. As the independent variables of the models, 59 functional groups were chosen based on the group contribution method. The prediction errors calculated from cross-validation were employed to determine the optimal parameters of two models. And, particle swarm optimization was used to get three parameters of SVM model. The PLS and SVM results of the average absolute errors for the whole data range from 58.59K and 29.11K, respectively, indicating that the predictive ability of the SVM is much superior than PLS.

• Fire Science and Engineering
• /
• v.15 no.2
• /
• pp.1-5
• /
• 2001

We had investigated combustion properties of starch. Decomposition of starch scavenged by pre-cipitator of spinning factory with temperature were investigated using DSC and TGA. Combustion properties of starch according to amount were checked as temperature variation according to time using spontaneous ignition apparatus. Moreover, combustion properties with blowing or without blowing condition were checked in spontaneous ignition apparatus. As results of thermal analyses, increase in raising temperature causes initial smoldering temperature to move towards low temperature section. In addition, as amount of starch was increased, initial smoldering temperature was lowered. All of combustion forms were smoldering combustion. Initial smoldering temperature was low more slightly with blowing condition than without blowing condition in spontaneous ignition apparatus, which condition made heating value high.

• Journal of the Korean Society of Safety
• /
• v.19 no.1
• /
• pp.137-140
• /
• 2004

The minimum ignition temperature at which the dust cloud can spontaneously ignite is considered to be very important in industries to prevent explosion occurring in hot surfaces. This paper has dealt with the experimental study of the determination of minimum ignition temperature of Hydroxy Propyl Methyl Cellulose (HPMC) dust cloud. We have used the Godbert-Greenwald Furnace Apparatus to determine the ignition temperature and limiting oxyten concentration for dust could. The experimental determinations on the minimum ignition temperature were carried out with various particle size with nominal diameters 45, 75 and 106${\mu}m$. The limiting oxygen concentration of dust cloud was determinated for the smaller size(45${\mu}m$) HPMC. Minimum ignition temperature of dust cloud was at 364$^{\circ}C$ for the concentration of 2.5g/L in the air and became higher with the increasing of nitrogen concentration. It was also found that the ignition didn't occur when the oxygen concentration was below 10%, and limiting oxygen concentration is at 11%.

• Proceedings of the Korea Institute of Fire Science and Engineering Conference
• /
• 2013.04a
• /
• pp.115-115
• /
• 2013

최근의 분진폭발은 플라스틱, 의약품, 목재, 곡물 저장고, 고체연료, 화학제품 제조공정 등을 포함하여 성형 및 가공 공정 등에서 화재폭발사고가 발생되고 있다. 폐목재를 재활용하여 PB(Particle board)를 생산하는 국내 제조사업장에서는 화재폭발 사고가 빈번히 발생하고 있어 예방대책이 요구되고 있다. 본 연구에서는 폐목재 제조공정의 사고예방과 목재분진 취급공정에 대한 안전대책 등을 제시하기 위하여 사고원인 물질인 폐목재 부유분진의 폭발특성실험을 실시하고 실험결과를 검토하였다. 또한 폐목재 분진의 화재폭발위험성을 상세히 평가하기 위하여 해당 물질의 자연발화점, 축열저장시험, 및 최소점화에너지 등의 화재폭발위험특성값을 실험적으로 조사하였다. 본 연구에서 사용한 폐목재 시료의 비구형 입자형태를 가지는데 입도분석기의 측정 결과 평균 입경은 $15.96{\mu}m$로 조사되었다. 또한 목재 분진의 함수율은 3.88%이며 중금속함유량은 1.73%이다. 자연발화점 측정결과 $225.5^{\circ}C$로서 비교적 낮게 측정되었고 퇴적분진에 대한 화재의 위험성은 높게 나타났다. 반면에 축열저장시험 결과를 통하여 공정관리 온도 및 보관온도를 $150^{\circ}C$ 이하로 관리하면 축열에 의한 자기분해 위험성은 낮은 것으로 판단되었다. 그러므로 축열에 의한 화재폭발 등의 위험성은 낮은 것으로 사료 된다. 최대폭발압력($P_{max}$)은 8.7 bar이며 폭발하한농도 (LEL)는 $60g/m^3$으로 나타났다. 부유분진의 폭발특성실험 결과 분진폭발지수(Kst)는 폭발등급 St 1 (0$bar{\cdot}m/s$)으로 나타났으며 폭발에 의한 위험성이 약한 분진으로 판정되었다. 최소점화에너지(MIE)는 10mJ < MIE <30mJ의 범위로 측정되었으며, 계산에 의해 추정된 최소점화 에너지(Es) 값은 14 mJ로서 일반적인 발화감도(Normal ignition sensitive)로 분류되었다. 이는 실질적인 점화원만 제거하여도 분진폭발을 예방할 수 있다는 것을 의미한다. 그러나 분진 폭발사고를 예방을 위하여 MIE값이 공정운전온도 $100^{\circ}C$ 초과 시에 급격히 낮아질 수 있으므로 운전 온도 설정에 있어서 주의가 필요하다.

• Fire Science and Engineering
• /
• v.27 no.6
• /
• pp.89-96
• /
• 2013

The purpose of this study is to check the fire origin and cause of the fire by observing the shape of the flame propagation in air-conditioner indoor unit through the fire tests. We supposed that the flame is spread from the surroundings to the top of the air-conditioner and ignited on top of the air-conditioner with n-heptane. And then repeated this experiment twice under the same conditions. After the ignition, refrigerant filled in air-conditioner and lubricating oil exploded with large explosion and flame having high temperature and pressure belched out rapidly due to bursting refrigerant pipe linked air-conditioner between 734 seconds and 559 seconds. After result of checking the combustion residue we found that almost all of that was lost except a part of the evaporator, motor and metal. The position of short-circuit traces of wiring for the air-conditioner ignited itself is similar to that of fire damage by external flame. Therefore, we verified that it is not certain to determine the ignition cause and point by only the shape of the combustion residues.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.1
• /
• pp.1-7
• /
• 2022

It has been found through physiological experiments that the visual neurons constituting the human visual cortex do not respond to all visual stimuli, but to a visual stimuli with specific conditions. In order to interpret such physiological experiments, a model that can simulate the firing characteristics of neurons including a linear filter with random gain was proposed. It has been proven through experiments that subspaces are formed. To verify the validity of the implemented model, the distribution of values for two pixels randomly extracted from four different visual stimulus data was observed. The difference between the two distributions was confirmed by extracting the central coordinate value, that is, the coordinate value with the most values, from the distribution of the total stimulus data and the spike ignition stimulus data. In the case of the entire set, it was verified through experiments that the stimulus data generating spikes is a subset or subspace of the entire stimulus data. This study can be used as a basic study related to the mechanism of spikes in response to visual stimuli.