• Analyses & Alternatives
• /
• v.1 no.1
• /
• pp.44-64
• /
• 2017

Russia's push in the Asia-Pacific region stems from its interests that have the national, regional and global dimensions. In their turn, the aims of this policy are civilizational, geopolitical, economic and prospective, with a long-term outlook. In the course of their achievement, cooperation with Northeast Asia's countries will play one of the key roles owing to the factors of geographic proximity, Northeast Asia's economic potential, risk hedging and a growing influence Northeast Asia exerts upon the global development. A new cooperation paradigm between Russia and the states of Northeast Asia should be based upon establishing and cementing self-reproducing ties. This is the central aim of Russian initiatives in relations, with Japan, the Republic of Korea, Democratic People's Republic of Korea, Mongolia and China. However, numerous obstacles ranging from Russia's absence in the regional free trade agreements and supply-production chains of value-added production to the permanent international instability generated by Pyongyang's missile-nuclear developments hamper the practical implementation of this task. Realizing the necessity to give an additional impetus to this new cooperative paradigm, Russia has to develop directions with an apparent consolidating effect. The most promising may be the establishment of a permanent security forum based upon Northeast Asia Peace and Security Mechanism chaired by the Russian Federation. The urgency of this measure and its expected support stem from the necessity to strengthen security in Northeast Asia, a task neither the US-led hub-and-spoke system nor ASE-AN-led multilateral dialogue venues have been able to resolve. The issues addressed at the security forum must include the unification of approaches to North Korean nuclear issue and producing a document specifying actions of the claimants on the disputed maritime territories in the "direct contact" situations. At the expert level, Russia has elaborated on the idea to establish such a forum outlining the spectrum of the key directions of cooperation. With the urgency in the establishment of this dialogue venue, its agenda has to be coordinated with the agendas of the existing security systems presented by the US alliances and the ASE-AN-led multilateral negotiations. The practical implementation of this initiative will strengthen security in Northeast Asia as its challenges will be resolved in the pre-emptive way based on coordinated approaches. Therefore, Russia as the Eurasian state will be one of the role players in the advent of the Asian century.

• Proceedings of the KIEE Conference
• /
• 2007.07a
• /
• pp.1777-1778
• /
• 2007

In this paper, We design and implement a humanoid robot, With Educational purpose, which can collaborate and communicate with human. We present an affective human-robot communication system for a humanoid robot, D2, which we designed to communicate with a human through dialogue. D2 communicates with humans by understanding and expressing emotion using facial expressions, voice, gestures and posture. Interaction between a human and a robot is made possible through our affective communication framework. The framework enables a robot to catch the emotional status of the user and to respond appropriately. As a result, the robot can engage in a natural dialogue with a human. According to the aim to be interacted with a human for voice, gestures and posture, the developed Educational humanoid robot consists of upper body, two arms, wheeled mobile platform and control hardware including vision and speech capability and various control boards such as motion control boards, signal processing board proceeding several types of sensors. Using the Educational humanoid robot D2, we have presented the successful demonstrations which consist of manipulation task with two arms, tracking objects using the vision system, and communication with human by the emotional interface, the synthesized speeches, and the recognition of speech commands.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.531-534
• /
• 2022

신경망 기반 심층학습 기술은 대화처리 분야에서 대폭적인 성능 개선을 가져왔다. 특히 GPT-2와 같은 대규모 사전학습 언어모델을 백본 네트워크로 하고 특정 도메인 타스크 대화 데이터에 대해서 미세조정 방식으로 생성되는 종단형 대화모델의 경우, 해당 도메인 타스크에 대해서 높은 성능을 내고 있다. 하지만 이런 연구들은 대부분 하나의 도메인에 대해서만 초점을 맞출 뿐 싱글 모델로 두 개 이상의 도메인을 고려하고 있지는 않다. 특히 순차적인 미세 조정은 이전에 학습된 도메인에 대해서는 catastrophic forgetting 문제를 발생시킴으로써 해당 도메인 타스크에 대한 성능 하락이 불가피하다. 본 논문에서는 이러한 문제를 해결하기 위하여 MultiWoz 목적지향 대화 데이터에 오픈 도메인 칫챗 대화턴을 유사도에 기반하여 추가하는 데이터 증강 방식을 통해 사용자 입력 및 문맥에 따라 MultiWoz 목적지향 대화와 오픈 도메인 칫챗 대화를 함께 생성할 수 있도록 하였다. 또한 목적지향 대화와 오픈 도메인 칫챗 대화가 혼합된 대화에서의 시스템 응답 생성 성능을 평가하기 위하여 오픈 도메인 칫챗 대화턴을 수작업으로 추가한 확장된 MultiWoz 평가셋을 구축하였다.

• Research in Mathematical Education
• /
• v.26 no.1
• /
• pp.1-17
• /
• 2023

Over the last three decades, there has been an increasingly strong emphasis on group-centered approaches to mathematics teaching. One primary responsibility for teachers who use group-centered instruction is to "check in", or intervene, with groups to monitor group learning and provide mathematical support when necessary. While prior research has contributed valuable insight for successful teacher interventions in mathematics group work, there is a need for more fine-grained analyses of interactions between teachers and students. In this study, we co-conducted research with an exemplary middle grade teacher (Ms. Green) to learn about fine-grained details of her intervention practices, hoping to generate knowledge about successful teacher interventions that can be expanded, replicated, and/or contradicted in other contexts. Analyzing Ms. Green's practices as an exemplary case, we found that she used exceptionally short interventions (35 seconds on average), provided space for student dialogue, and applied four distinct strategies to support groups to make mathematical progress: (1) observing/listening before speaking; (2) using a combination of social and analytic scaffolds; (3) redirecting students to task instructions; (4) abruptly walking away. These findings imply that successful interventions may be characterized by brevity, shared dialogue between the teacher and students, and distinct (and sometimes unnatural) teaching moves.

• Annual Conference on Human and Language Technology
• /
• 2024.10a
• /
• pp.566-569
• /
• 2024

본 논문에서는 지식 기반 대화 Task의 특성을 반영하는 LLM 기반 자동 평가 방법을 제안한다. 자연어 처리에서 생성된 텍스트를 자동으로 평가하는 데에는 의미 이해, 대화 맥락 파악 등의 어려움이 존재했지만, 대규모 언어모델(Large Language Model, LLM)의 등장으로 텍스트의 자동 평가 품질이 개선되었다. 이에 LLM과 Prompt Engineering의 연구가 급속도로 진행되고 있으며, Task의 특성을 분석해, 프롬프트에 반영하는 것이 큰 연구 방향이 되고 있다. 특히, 기존의 텍스트 생성 Task와는 달리 지식기반 대화의 평가에서는 신뢰성을 확보하기 위해 고려해야 할 세 가지 주요 특성이 존재한다. 이 연구는 그 특성을 반영한 프롬프트를 통해 LLM 기반 자동 평가 진행 후 사람평가와 유사한 결과를 도출할 수 있음을 확인한다.

• KIPS Transactions on Software and Data Engineering
• /
• v.8 no.5
• /
• pp.213-222
• /
• 2019

Determining the meaning of a keyword in a speech dialogue system is an important technology for the future implementation of an intelligent speech dialogue interface. After extracting keywords to grasp intention from user's utterance, the intention of utterance is determined by using the semantic mark of keyword. One keyword can have several semantic marks, and we regard the task of attaching the correct semantic mark to the user's intentions on these keyword as a problem of word sense disambiguation. In this study, about 23% of all keywords in the corpus is manually tagged to build a semantic mark dictionary, a synonym dictionary, and a context vector dictionary, and then the remaining 77% of all keywords is automatically tagged. The semantic mark of a keyword is determined by calculating the context vector similarity from the context vector dictionary. For an unregistered keyword, the semantic mark of the most similar keyword is attached using a synonym dictionary. We compare the performance of the system with manually constructed training set and semi-automatically expanded training set by selecting 3 high-frequency keywords and 3 low-frequency keywords in the corpus. In experiments, we obtained accuracy of 54.4% with manually constructed training set and 50.0% with semi-automatically expanded training set.

• Journal of the Korean Regional Science Association
• /
• v.39 no.4
• /
• pp.73-90
• /
• 2023

This research examines the impact of robotics integration on job dynamics in the U.S. manufacturing sector, adding to the critical dialogue on technological evolution and the future of jobs. Anchored in the task-model framework, the study hypothesizes that robotic integration exerts differential influences on diverse occupational clusters, each identified by their unique task-specific attributes. An in-depth examination was undertaken to elucidate the interplay between robotic integration and the occupation clusters. Employing a multilevel growth curve model, our empirical investigation tracked employment dynamics from 2012 to 2022 across 52 U.S. regions, covering 307 manufacturing occupations. The findings suggest a pronounced job decline within occupations necessitating manual dexterity. Nonetheless, the evidence does not conclusively support that the extent of robotics integration exacerbates this trend. These findings imply that the employment shifts in the U.S. manufacturing sector are predominantly driven by long-standing trends of deindustrialization and functional specialization, rather than by the recent diffusion of robotic technologies.

• Korean Journal of Child Studies
• /
• v.18 no.2
• /
• pp.311-331
• /
• 1997

This study investigated the relationship between preschool children's social competence and perceived social support. The subjects were 85 children enrolled in 10 kindergartens and nurseries located in Seoul. Social support was assessed by children themselves with a social support dialogue Cask, "My Family and Friends," developed by Reid, Lamdegman, and Jaccard (1989). This task consists of 12 dialogues based on Vygotskian principles. Social competency was assessed by teachers with a slightly modified version of Song's Iowa Social Competency Scale (ISCS). The data were analyzed by frequencies percentiles, mean, one-way ANOVAs, and Pearson's correlations. Results showed that (1) Children perceived parents to be the best multi-purpose social providers: mothers for emotional support and fathers for instrumental support. (2) Children with unemployed mothers perceived higher maternal emotional, informational, and instrumental support than children with employed mothers. (3) Middle-class children perceived higher friendship support than upper-or lower-class children. (4) Friendship support was the one variable most significantly related to children's social competence.

• Korean Journal of Child Studies
• /
• v.25 no.3
• /
• pp.27-39
• /
• 2004

This study was to find out children's Negotiation Levels(NLs) according to their age and gender. The subjects for this study were 143 children of 7, 9, 12 years. Two children dyads participated in the experimental play situation with Rokenbok Electronic Toy Systems. The NLs in children's interactive dialogue & actions were scored with Stone, Robinson & Taylor(1980)'s 'Negotiation of Task Completion Coding Manual'. They were coded into Level 0(no interaction), Level l(one way interaction), Level 2(reciprocally interaction), Level 3(mutual or cooperative interaction). The results were as follows; 1) 12-year-old children negotiated with higher level than 7-and 9-year old children 2) Boys' NLs average were higher than girls'. These results imply that 1) the fluctuation of NLs in childhood, especially around 10 years, should be interpreted carefully, 2) various measuring kits for negotiation should be developed considering children's characteristics, such as age, gender etc.

• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.246-253
• /
• 2006

대화 에이전트와 관련된 지금까지의 연구는 대개 대상 도메인을 한정하고, 특정 목적을 달성하기 위해 사용자와 대화할 수 있는 에이전트에 관한 연구가 많았다. 본 연구에서는 도메인이 한정되지 않은 일반 도메인 대화에서 화행(speech act)정보를 수동으로 부착시켜 구축한 말뭉치에 대해 소개하고 이 말뭉치를 토대로 자동으로 화행을 분류할 수 있는 유용한 자질들을 선보인다. 그리고 도메인이 한정된 말뭉치와 도메인이 한정되지 않은 말뭉치를 자동으로 화행분류해 본 실험한 결과를 비교하였다.