• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.219-240
• /
• 2023

Conversational AI which allows users to interact with satisfaction is a long-standing research topic. To develop conversational AI, it is necessary to build training data that reflects real conversations between people, but current Korean datasets are not in question-answer format or use honorifics, making it difficult for users to feel closeness. In this paper, we propose a conversation dataset (KOMUChat) consisting of 30,767 question-answer sentence pairs collected from online communities. The question-answer pairs were collected from post titles and first comments of love and relationship counsel boards used by men and women. In addition, we removed abuse records through automatic and manual cleansing to build high quality dataset. To verify the validity of KOMUChat, we compared and analyzed the result of generative language model learning KOMUChat and benchmark dataset. The results showed that our dataset outperformed the benchmark dataset in terms of answer appropriateness, user satisfaction, and fulfillment of conversational AI goals. The dataset is the largest open-source single turn text data presented so far and it has the significance of building a more friendly Korean dataset by reflecting the text styles of the online community.

• Annual Conference on Human and Language Technology
• /
• 1997.10a
• /
• pp.271-276
• /
• 1997

대화체는 문어체와는 달리 화자와 청자 사이의 질의/응답으로 이루어진 형태의 문장들을 가지며, 생략과 대용어가 빈번히 발생하는 특징을 갖는다. 이러한 대화 형태에서 어떠한 한 문장에는 화자가 전달하고자 하는 의도를 포함하고 있다. 이러한 대화체 문장들을 번역하는 것은 단순한 언어적 분석에 의한 번역으로서는 많은 번역상의 오류가 발생하게 된다. 따라서 대화체 문장들의 올바른 번역을 위해서는 대화의 상황을 반영하는 문맥 정보가 부가적으로 요구된다. 본 연구에서는 이러한 문맥 정보로서 화행을 사용하여 대화체 기계번역을 수행하고자 한다. 화행(Speech Act)이란 화자에 의해 의도되어 발화 속에 포함된 언어적 행위를 나타내며, 이러한 화행을 분석함으로써 화자의 의도를 파악하고 이를 통해 올바른 번역을 수행할 수 있게 된다. 본 기계번역 시스템에 포함된 화행 분석 과정에서는 대화를 화행으로 모델링한 담화 문법과 유사한 형태의 재귀적 대화 전이망(Recursive Dialog Transition Network)을 사용하게 된다. 본 논문에서는 호텔 예약 영역에서의 기계번역 시스템에 대한 간단한 소개와 화행의 종류 및 분석 방법과 이를 통한 기계번역 방식에 대해 살펴보도록 하겠다.

• Journal of Multimedia Information System
• /
• v.7 no.2
• /
• pp.141-146
• /
• 2020

Cine-Hangeul is a program that can predict the running time of a movie based on the screenplay before production. This paper seeks to verify the prediction reporting function of Cine-Hangeul, which is the standard Korean screenplay format. Moreover, this paper presents a method to increase the accuracy of the Cine-Hangeul reporting function. The objective of this paper is to offer a correction method based on scientific evidence because the current Cine-Hangeul reporting function has many errors. The verification process for five scenarios and movies confirmed that the default setting value of Cine- Hangeul's screening time prediction reporting was many errors. Cine-Hangeul analyzes the amount of textual information to predict the time of the scene and the time of the dialogue and helps predict the total time of the movie. Therefore, if a certain amount of text information is not available, the accuracy is unreliable. The current Cine-Hangeul prediction report confirms that the efficiency is high when the scenario volume is about 90 to 100 pages. As a result, prediction of screening time by Cine-Hangeul, a Korean scenario standard format program, confirmed the verification that it could secure the same level of reliability as the actual screening time by correcting the reporting settings. This verification also affirms that when applying about 50 percent of the basic set of screening time reporting, it is almost identical to the screening time.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.204-208
• /
• 2007

대화시스템은 사용자가 자연언어를 사용하여 해당 시스템과 필요한 정보를 주고받는 목적 지향적 에이전트로서 활용되어 왔다. 이러한 대화형 에이전트는 사용자의 입력으로부터 필요한 정보를 정확하게 추출함으로써 이후 처리단계에서의 결과를 향상시킬 수 있다. 본 논문에서는 항공여행관련 대화에서 발생하는 예약정보들 중에서 경유정보, 특히 경유하는 시간 및 날짜에 대한 정보를 효과적으로 추출하는 방법에 대해서 다룬다. 출발 도착정보와 경유정보를 계층적으로 분류하고, 현재 발화되고 있는 문장보다 선행되고 있는 문장들의 예약정보들을 문맥정보로 사용하여 현재 문장에서 추출하고자 하는 정보들을 학습하고 평가하였다. 이를 통해서 얻어진 결과는 출발.도착 및 경유정보를 동시에 고려했을 때보다 효과적인 학습 성능을 보였으며 실제로 시간정보에 대해서는 81.5%, 날짜정보에 대해서는 92.0%의 정확도를 보였다.

• Korean Journal of Cognitive Science
• /
• v.15 no.3
• /
• pp.35-44
• /
• 2004

As the use of internet becomes proliferated, a great amount of information is provided through diverse channels. Users require effective information providing service and we have studied the conversational agent that exchanges information between users and agents using natural language dialogue. In this paper, we develop a question answering agent providing the corresponding answer by analyzing the user's intention using artificial intelligence techniques such as pattern matching and Bayesian network We work out various problems in knowledge representation of users by constructing keyword synonym database. The proposed method is applied to designing an agent for the introduction of a fashion web site, which confirms that it responds more flexibly to the user's queries.

• Journal of Magnetics
• /
• v.11 no.4
• /
• pp.182-188
• /
• 2006

The search for high-density recording materials has been one of most active and vigorous field in the field of magnetism. $FePt-L1_{0}$ nanoparticle has emerged as a potential candidate because of its high anisotropy. In this paper, we provide an overview of recent work at Argonne National Laboratory that contributes to the ongoing dialogue concerning the relation between structure and properties of the FePt nanoparticle system. In particular we discuss the ability to control structure and properties via dosing with Cr. Cr-dosed FePt films were grown via molecular beam epitaxy and annealed at $550^{\circ}C$ in an ultrahigh vacuum chamber, and were studied with the surface magneto-optic Kerr effect (SMOKE), scanning electron microscopy (SEM) and x-ray magnetic circular dichroism (XMCD). We found that small dosage of Cr helps to generate $L1_{0}$ phase FePt magnetic nanoparticles with small size, defined shape and regular spatial distribution on MgO (001) substrate. The nanostructures are ferromagnetic with high magnetic coercivity (${\sim}0.9T$) and magnetic easy axis in the desired out-of-plane orientation. We also show that controlling the lateral region where nanostructures exist is possible via artificial patterning with Cr.

• Research in Community and Public Health Nursing
• /
• v.12 no.2
• /
• pp.317-328
• /
• 2001

The purposes of this study were to develop the empowerment education program, to describe the experiences of diabetes patient's empowerment process and to develop a theoretical model of the diabetes patient's empowerment process. Method 1. : The development of the empowerment program for the diabetes patients: The strategies of the empowerment education program were enhancement of problem - solving, decision making, self-efficacy, self-control. participation and mutual support. Method 2. : According to the grounded theory methodology of Strauss and Corbin, the qualitative data was collected with in depth interviews and participants observations until its saturation when the 25 consented subjects were participating and interacting with the other subjects in the empowerment education program. Results: With the analysis of the data, 29 categories were generated. The core category generated, which was a central phenomenon of the empowerment process, was named powerlessness. The intervening conditions facilitating or impeding the empowerment process were discovered as supportive systems through the participation of group meeting, problem solving dialogue, and the knowledge deficit of self-care. The action/interaction strategies were developed as the paricipating, dialoguing, questioning, supporting system, self-controlling, self efficacy, enhancing self-esteem. stress relaxing and instillation of hope.

• Annual Conference on Human and Language Technology
• /
• 1994.11a
• /
• pp.313-317
• /
• 1994

본 연구에서는 인간과 컴퓨터 사이의 음성을 이용한 대화 시스템을 구현하였다. 특별히 음성을 인식하는데 있어서 단어추출(word apotting) 방법을 사용하는 경우에 알맞은 의미 분석 방법과 도표 형태의 규칙을 기반으로 하여 시스템의 응답을 생성하는 방법에 대하여 연구하였다. 단어추출 방법을 사용하여 음성을 인식하는 경우에는 형태소분석 및 구문분석의 과정을 이용하여 사용자의 발화 의도를 분석하기 어려우므로 새로운 의미분석 방법을 필요로 한다. 본 연구에서는 퍼지 관계를 사용하여 사용자의 발화 의도를 파악하는 새로운 의미분석 방법을 제안하였다. 그리고, 사용자의 발화 의도에 적절한 시스템의 응답을 만들고 응답의 내용을 효율적으로 관리하기 위한 방범으로 현재의 상태와 사용자의 의도에 따른 응답 규칙을 만들었다. 이 규칙은 도표의 형태로 구현되어 규칙의 갱신 및 확장을 편리하게 만들었다. 대화의 영역은 열차 예매에 관련된 예매, 취소, 문의 및 관광지 안내로 제안하였다. 음성의 오인식에 의한 오류에 적절히 대처하기 위해 시스템의 응답은 확인 및 수정 과정을 포함하고 있다. 본 시스템은 문자 입력과 음성 입력으로 각각 실험한 결과, 사용자는 시스템의 도움을 받아 자신이 의도하는 목적을 달성할 수 있었다.

• Journal of Digital Convergence
• /
• v.10 no.7
• /
• pp.253-259
• /
• 2012

In this study, the dialog syntax-based video content production flow for the character set, 'Form Noah' chart using the mouse, lip-sync Animation been making 3D characters were applying. Vertex Animation Morphing techniques by expressing the natural shape of the mouth for the characters engaging and the transmission of visual information for the viewers to be able to get a high intelligibility is considered.

• Speech Sciences
• /
• v.9 no.2
• /
• pp.25-47
• /
• 2002

Human language activity in the information society has been developing the communication system between humans and machines. The aim of this study was to analyze dialectal speech in Korea. One hundred Kyungsang and one hundred Cholla informants participated in this study. A CSL and Flexible laryngeal fiberscope were used for analysis of the acoustic and glottal gestures of all the vowels and consonants. Test words were made on the picture cards and letter cards which contained each vowel and each consonant, respectively. The dialogue between the examiner and the informants was recorded in a question and answer manner. The acoustic results of two dialects were as follows: Kyungsang and Cholla informants showed neutralization between /e/ and /$\varepsilon$. However, the apertures of Kyungsang vowels /i, w, u, o/ were higher than those of Cholla vowels. The /wi/ and /$\varepsilon$/ of Kyungsang Diphthong vowels were shown as simple vowels /i/ and /$\varepsilon$/ in Cholla dialect. The VOT of Cholla dilaect was longer than that of Kyungsang dialect. The fricative frequence of Kyurlgsang dialect was about 1000Hz higher than that of Cholla dialect. The glottal widths on fiberscopic images showed that the consonant durations of Kyungsang and Cholla dialects were correlated all together with the acoustic duration on the spectrogram.