• Journal of Power System Engineering
• /
• v.6 no.3
• /
• pp.5-10
• /
• 2002

• Korean Journal of Cognitive Science
• /
• v.17 no.2
• /
• pp.139-150
• /
• 2006

Speech recognition errors cause fatal results in a spoken dialogue system. When a system can not determine the speech-act of u utterance due to speech recognition errors, a dialogue system has a difficulty in continuing conversation. In this paper, we propose strategies for sub-dialogue generation by inferring the speech-act of an utterance with patterns of recognition errors on the field of form-filling dialogue. We used the proposed method on a plan-based dialogue model, corrected 27% of incomplete tasks, and acquired overall 89% of task completion rate.

• Annual Conference on Human and Language Technology
• /
• 1998.10c
• /
• pp.370-375
• /
• 1998

본 논문에서는 화행처리를 기반으로 한 중고자동차 매매 영역 대화시스템 구축 과정에서 활용한 자연언어 대화체 발화의 언표적 행위와 언표내적 행위 간의 관계를 살펴보고자 한다. 하나의 언표적 행위는 여러 상이한 언표내적 행위를 수행할 수 있으며 또 역으로 여러 언표적 행위가 동일한 언표내적 행위를 구성하기도 한다. 이처럼 언표적 행위와 언표내적 행위 간에는 획일적이고 일률적인 상관관계를 설정하기가 어렵다. 그럼에도 불구하고 대화에 나타나는 발화들은 어떤 행위를 구성하는 경향이 강하고 이 행위들은 각 발화의 언표내적 행위로 결정되며 단일 발화들이 모여서 이루는 대화에는 화행의 흐름이 있기 마련이다. 따라서 우리는 본 논문에서 대화시스템의 효율적인 대화관리를 위해서 실제 자연언어 발화와 그에 따른 언표적 행위, 언표내적 행위를 올바로 분석하고 그 관계를 규명하여 대화를 자연스럽게 유도하도록 한다.

• Cross-Cultural Studies
• /
• v.27
• /
• pp.435-470
• /
• 2012

This paper aims to examine directives and its adjacency pair directives-responses in the Chinese textbook Hanyu Kouyu from the speech act theory, and give some suggestions for Chinese speaking instruction. For analysis of directives, it is important to consider context surrounding it at first. Context includes space for speaking, person who speak with, adjacent speech act, etc., so this paper considers context firstly before explaining directives and discusses how each element of context effects interpretation of its meaning. From the politeness perspective, directives and rejection as one of the responses are easy to be FTAs(face-threatening acts) fundamentally, so they are required to have some strategies for minimizing the threat, and it is found that there are strategies for positive face and negative face in the textbook. In the textbook, it is also examined that directives-responses contribute for interlocutor to accomplish social acts, for example, negotiating, insisting, maintaining, making alternative ideas, etc.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2001.11a
• /
• pp.73-74
• /
• 2001

• Journal of KIISE:Software and Applications
• /
• v.35 no.5
• /
• pp.306-312
• /
• 2008

In a goal-oriented dialogue, speaker's intention can be approximated by a semantic structure that consists of a pair of a speech act and a concept sequence. Therefore, it is very important to correctly identify the semantic structure of an utterance for implementing an intelligent dialogue system. In this paper, we propose a model to efficiently analyze the semantic structures based on an active teaming method. To reduce the burdens of high-level linguistic analysis, the proposed model only uses morphological features and previous semantic structures as input features. To improve the precisions of semantic structure analysis, the proposed model adopts CRFs(Conditional Random Fields), which show high performances in natural language processing, as an underlying statistical model. In the experiments in a schedule arrangement domain, we found that the proposed model shows similar performances(92.4% in speech act analysis and 89.8% in concept sequence analysis) to the previous models although it uses about a third of training data.

• Journal of Korean language education
• /
• v.29 no.4
• /
• pp.225-254
• /
• 2018

The purpose of this study is to contrast the patterns of realization and understanding of refusal speech acts between Korean and Thai learners. This study intends to answer the following questions: (1) Do Koreans and Thai learners perform refusal speech acts differently? (2) Do Koreans and Thai learners understand refusal speech acts differently? A DCT and a follow-up interview were conducted to collect data of two groups of 30 native Korean speakers and 30 native Thai speakers. For research question 1, we analyzed the refusal strategy and provided reasons given by Koreans and Thai learners depending on the context. For research question 2, we ran a chi-squared test on the elements of the follow-up interviews, such as the weight of burden of refusing, and whether the participant would actually refuse or not. The differences between the refusal strategies of the two groups could be categorized by the preceding inducing speech act. In refusing a request, the difference was prominent in the apologizing strategy, whereas in refusing a suggestion, the difference was mainly in the direct refusal strategy. When refusing an invitation, the most evident difference was the number of refusal strategies employed. When providing an explanation of refusal to people with high social status, Koreans gave more specific reasons for refusals, whereas Thai learners tended to use more vague reasons. Moreover, when refusing an invitation, Koreans primarily mentioned the relationship, and Thai learners showed the spirit of Greng Jai. When asked the weight of burden of refusing, Koreans felt pressured to refuse a request from people with high social status, and a suggestion or invitation from people with high level of intimacy while Thai learners found it highly difficult to make a refusal in all cases. In answering whether they would actually refuse or not, Koreans tried not to make a refusal to people with high level of intimacy, and such a trend was not evident among the Thai. This study can help us better understand the learner's pragmatic failure, and serve as a basis in establishing a curriculum for teaching speech acts.

• Proceedings of the KOR-KST Conference
• /
• 1995.05a
• /
• pp.233-236
• /
• 1995

봉상해성시공공교통적굉함관리,주요시제정정초적장당정책화기아정책, 가강수구관리화행당관리.기내용시,성시객리교통이방전공공교통편강선,공공교통극이건고대용량적쾌속후도교통위중점.합리성시포국,조린거민승행,인부교 통수구.평형리최리능.하요도성시객리행당관행집중축일,유효유서적관리.

• Korean Journal of Cognitive Science
• /
• v.20 no.4
• /
• pp.421-437
• /
• 2009

A chatting system is a computer program that simulates conversations between a human and a computer using natural language. In this paper, we propose a statistical model to generate natural chatting sentences when keywords and speech acts are input. The proposed model first finds Eojeols (Korean spacing units) including input keywords from a corpus, and generate sentence candidates by using appearance information and syntactic information of Eojeols surrounding the found Eojeols. Then, the proposed model selects one among the sentence candidates by using a language model based on speech act information, co-occurrence information between Eojeols, and syntactic information of each Eojeol. In the experiment, the proposed model showed the better correct sentence generation rate of 86.2% than a previous conventional model based on a simple language model.

• Journal of Korean language education
• /
• v.25 no.1
• /
• pp.83-113
• /
• 2014

This study aims to examine the change of strategy use by analyzing the refusal sequence of beginner Japanese Korean learners by time difference of 10 weeks. The results of the study are as follows; First, in the solidarity system of politeness(power-, distance-), learners showed similar ability as native Koreans after 10 weeks. Second, in the deference system(power-, distance+), there was a difference according to the situational burden when using strategies. In the invitational setting, the learners sought justification for their refusal or tried to negotiate by explaining in detail as time passed. However, in the request setting, they showed no will to compromise by refusing directly and leading the listener to give up. Third, similar difference was found in the use of strategies depending on the situational burden in the hierarchical system(power+, distance+). While learners could present their justification for refusal by showing their interest in the invitation as time passed in the invitation setting, they expressed straight refusal and were not willing to negotiate in the request setting.