• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.518-523
• /
• 2021

챗봇과 같은 대화형 에이전트 사용이 증가하면서 채팅에서의 혐오 표현 사용도 더불어 증가하고 있다. 혐오 표현을 자동으로 탐지하려는 노력은 다양하게 시도되어 왔으나, 챗봇 데이터를 대상으로 한 혐오 표현 탐지 연구는 여전히 부족한 실정이다. 이 연구는 혐오 표현을 포함한 챗봇-사용자 대화 데이터 35만 개에 한국어 말뭉치로 학습된 KoELETRA 기반 혐오 탐지 모델을 적용하여, 챗봇-사람 데이터셋에서의 혐오 표현 탐지의 성능과 한계점을 검토하였다. KoELECTRA 혐오 표현 분류 모델은 챗봇 데이터셋에 대해 가중 평균 F1-score 0.66의 성능을 보였으며, 오탈자에 대한 취약성, 맥락 미반영으로 인한 편향 강화, 가용한 데이터의 정확도 문제가 주요한 한계로 포착되었다. 이 연구에서는 실험 결과에 기반해 성능 향상을 위한 방향성을 제시한다.

• Cartoon and Animation Studies
• /
• s.37
• /
• pp.1-24
• /
• 2014

The advancement of computer technology has blurred the border in image expression between live-action films and 3D animation to an indistinguishable extent. Despite the ever-evolving image expression, it has been found that animation properly incorporating abstractive expression receives more positive responses from the audience than the animation closely akin to live-action films. This indicates that animation-specific abstractive expression can exert more significant effects than that of live-action expression as proved in Dr. Mori Masahiro's 'Uncanny Valley.' Also, such effects apply only to characters, whilst the technology capable of live-action expression is acknowledged only for the background of 3D animation. Without doubt, animation has its own infinite world of expression, and thus more effective methods for its expression can no longer be considered a matter of technology. The present study analyzes the abstractive expression of 3D animation in terms of characters and backgrounds and explores a more effective method to apply the current technology of image expression to 3D animation with a view to proposing a direction for more appropriate application of 3D animation technology to future animation.

• Korean Journal of Cognitive Science
• /
• v.10 no.4
• /
• pp.71-83
• /
• 1999

Conventional symbolic inference systems lack flexibility because they do not well reflect flexible semantic structure of knowledge and use symbolic logic for their basic inference mechanism. For solving this problem. we have recently proposed the 'Connectionist Semantic Network(CSN)' as a model for flexible knowledge representation and inference based on neural networks. The CSN is capable of carrying out both approximate reasoning and commonsense reasoning based on similarity and association. However. we have difficulties in representing general and structured high-level knowledge and variable binding using the connectionist framework of the CSN. In this paper. we propose a hybrid system called SymCSN(Symbolic CSN) that combines a symbolic module for representing general and structured high-level knowledge and a connectionist module for representing and learning low-level semantic structure Simulation results show that the SymCSN is a plausible model for human-like flexible knowledge representation and inference.

• Journal of Korea Multimedia Society
• /
• v.8 no.4
• /
• pp.566-574
• /
• 2005

In spite of recent increase in the use of avatar in Web and Virtual Reality, there has not been a service that allows users to control directly the avatar behaviors. In addition, the conventional behavior control languages required a lot of complicated information for controlling the avatar motions. Moreover, in order to apply written languages to a different task domain, it was necessary to modify or rewrite the languages. In this paper, we define Task-Level Behavior Description Language and Motion Representation Language for more simple control of the avatar behavior. The first thing allows describing the avatar behaviors in each task domain, and The second thing enables writing detailed data for motion control. And in this paper, we developed an interpreter which can automatically change the Behavior Description Language to the Motion Representation Language. So this system allow users control the avatar behavior simply with only use the Behavior Description Language. The system was applied to shopping mall and the Task-level Behavior Description Language was compared with conventional languages to see how it was more effective in behavior description.

• The Journal of the Korea Contents Association
• /
• v.18 no.3
• /
• pp.409-420
• /
• 2018

The decision to actively reveal oneself as a consumer on social media depends on whether the focus is more on positive rewards than negative threats associated with self-presentation, that is, who the other party is in that relationship. The audience of self-presentation could also be divided into expressive and instrumental ties: work partners can be classified as an audience of instrumental ties, whereas school friends can be regarded as an audience of expressive ties. This study accordingly predicted that people would focus more on prevention of negative threats than positive reward, exhibiting defensive self-presentation in relationships with strong instrumental ties because the fallout from negative threats due to failed self-presentation have a more detrimental effect than a positive reward from successful self-presentation. The empirical findings thereby indicated that as the proportion of coworkers increased among Facebook friends, the mediation effect through defensive self-presentation increased, whereas such effect through acquisitive self-presentation decreased.

• School Mathematics
• /
• v.17 no.3
• /
• pp.407-421
• /
• 2015

Linear system is a basic subject matter of school mathematics courses. Even though elimination is a useful method to solve linear systems, its fundamental principles were not discussed pedagogically. The purpose of this study is to help the development of mathematical content knowledge on linear systems conceptions. To do this, various representations and translations among them were considered, and in particular, the basic principles for elimination method are analyzed geometrically. Rectangular representation is used to solve word problem treated in numbers of things in elementary mathematics and it is useful as a pre-stage to introduce elimination. Slopes and intercepts of lines associated linear equations are used to obtain the Cramer's formula and this solving method was showing the connection between algebraic and geometric procedures. Strategy deleting variables of linear systems by elementary operations is explored and associated with the movements of lines in the family of lines passing through a fixed point. The development of mathematical content knowledge is expected to enhance pedagogical content knowledges.

• Journal of Korean Library and Information Science Society
• /
• v.51 no.1
• /
• pp.1-20
• /
• 2020

This study is to find the methods that BIBFRAME could accept the new concept of representative expression in the RDA Toolkit Beta which has developed in 2019. Research methods were the literature reviews and the mapping between RDA Toolkit Beta and BIBFRAME. BIBFRAME's acceptance of representative expression of RDA Toolkit Beta was suggested as followings. First, the properties of representative expression should be defined in BIBFRAME because the RDA Toolkit Beta defined the expression attributes and representative expression elements apart. Three choices for BIBFRAME's acceptances are (1) to develop the newly devised representative expression properties (2) to specify representative expression properties with using refinement (3) to differentiate representative expression properties with using class. Second, as relationship elements should be defined in BIBFRAME to link work and representative expression which transfer the expression attribute to work. This study will contribute to revise the BIBFRAME because this focusd on BIBFRAME's acceptance of RDA Toolkit Beta in which reflected LRM.

• Education of Primary School Mathematics
• /
• v.12 no.2
• /
• pp.81-97
• /
• 2009

Visual representations play an important role for students to understand the meaning of a given problem, devise problem-solving approaches, and implement them successfully. The purpose of this study was to investigate how 6th graders would use visual representations in solving mathematical problems and in what ways such use might affect successful problem solving. The results showed that many students preferred numerical expressions to visual representations. However, students who used visual representations, specifically schematic representations, performed better than those who employed numerical representations. Given this, this paper includes instructional implications to nurture students' use of visual representations in a way to increase their problem solving ability.

• The Journal of Information Technology and Database
• /
• v.5 no.1
• /
• pp.13-37
• /
• 1998

인터넷 등의 컴퓨터 통신 네트워크의 발달로 인하여 분산된 정보자원의 공유를 통한 자원에 대한 재사용성의 필요성이 대두되었다. IRD(Information Resource Dictionary)는 조직 내에서 관련된 모든 정보에 대한 데이터가 논리적으로 중앙화된 정보저장소(repository)이다. IRD 내의 데이터는 다른 데이터를 기술하므로 이른바 메타 데이터라고 하기도 한다. IRD의 사전(dictionary) 요소는 정보자원의 종류, 정보자원의 의미, 정보자원의 논리적 구조, 정보자원의 위치, 그리고 정보자원의 접근방법 등을 기술한다. FIPS ANSI의 IRDS는 이항 관계를 이용하여 무결성 제약조건을 표현하므로 제약조건 규칙의 표현과 일반적인 추론 규칙의 표현이 제한되어 있으며, 다양한 형태의 무결성 제약조건의 표현과 IRD와 관련된 여러 정보의 도출 또는 추론 및 관리에 관한 사항은 IRD 응용 고유의 문제로 간주하여 언급하고 있지 않다. 한편, FIPS IRDS는 사용자가 SQL 및 IRD에 대한 전문적 지식이 없이는 사용자 질의 작성이 어려운 점등에 대한 문제점을 안고 있다. 본 논문은 FIPS IRDS의 기본모델에서 정보자원 표현, 정보자원들간의 관계, 정보자원의 관리 정보 구분을 명확히 하기 위해 정보자원 모델을 정보자원 표현요소와 정보자원 관리요소의 두 부류로 나누어 구분하고, 각 부류에 대한 자격 질의(competency question)를 통하여 유추된 요소들을 FIPS ANSI IRDS 기본 모델의 스키마 기술 레벨과 스키마 레벨에 첨가함으로써 그 기본 모델을 확장한다. 그리고, FIPS ANSI IRDS가 제공하는 IRD 기술과 관리 기능을 그대로 포함하면서 앞에서 문제점으로 지적된 제약조건 표현과 추론규칙 표현을 위하여 확장된 기본 모델을 중심으로 각 레벨의 구성 요소들의 형식적 의미(formal semantics)와 레벨 내 혹은 레벨 구성요소들간의 관계성(relationship), 그리고 제약조건의 표현과 질의 추론 규칙들을 식별하여 FOPL(First Order Predicate Logic)로 표현한다. 또한, 본 논문은 FOPL로 표현된 predicate들과 규칙들을 구현하기 위하여 Prolog로 변환하기 위한 이론적 방법론을 제시하고 정보자원 관리를 위한 기본 함수들과 스키마 진화(schema evolution)를 위한 방법론을 제안한다.

• Korean Journal of Health Psychology
• /
• v.15 no.3
• /
• pp.483-498
• /
• 2010

This study is to examine the effects of emotion regulation strategies in facial expressions and emotional experiences, based on the facial expressions of groups, using antecedent- and response- focused regulation. 50 female undergraduate students were instructed to use different emotion regulation strategies during the viewing of a disgust inducing film. While watching, their facial expressions and emotional experiences were measured. As a result, participants showed the highest frequency of action units related to disgust in the EG(expression group), and they reported in the following order of DG(expressive dissonance group), CG(cognitive reappraisal group), and SG(expressive suppression group). Also, the upper region of the face reflected real emotions. In this region, the frequency of action units related to disgust were lower in the CG than in the EG or DG. The results of the PANAS indicated the largest decrease of positive emotions reported in the DG, but an increase of positive emotions reported in the CG. This study suggests that cognitive reappraisal to an event is a more functional emotion regulation strategy compared to other strategies related to facial expression and emotional experience that affect emotion regulation strategies.