• The Journal of the Korea Contents Association
• /
• v.20 no.10
• /
• pp.284-293
• /
• 2020

The rapid progress of AI technology gives us the expectation to solutions to various problems in our society, and at the same time, it gives us anxiety about the side effects that can occur if AI develops beyond human control. This study was conducted in the early 20s with less objection to advanced devices. We attempted to provide clues to understand thoughts and attitudes of the targets about the future environment that will be brought by AI through the process of finding intent the acceptance of strong AI technology. For this, we applied the Theory of Planned Behavior, and further expanded this research model to identify factors affecting the attitude toward AI. As a result, the attitude toward AI and perceived behavioral control had a significant effect on the intention to use to strong AI. In addition, we found that the expectation of the benefit of improving task performance and the anxiety on the threat of relationship disturbance had a significant effect on the attitude toward AI. This study suggests implications for AI-related companies establishing the direction of technology development and for government setting a policy direction for AI adoption.

• Journal of Korea Society of Industrial Information Systems
• /
• v.13 no.3
• /
• pp.38-53
• /
• 2008

This study focuses on identifying motivation factors affecting knowledge sharing behavior in virtual communities. To achieve this research objective, a research model has been developed through two major coordinated theories such as social cognitive theory(self-efficacy and community-related outcome expectation) and social exchange theory (codification effort, image, reciprocity, and enjoyment in helping others). 246 samples have been collected and simple regression and multiple regression methods have been used for empirical analysis. The research result is that self-efficacy has a positive influence on image, reciprocity, enjoyment in helping others, and community-related outcome expectation and thus this result reveals that self-efficacy indirectly affects knowledge sharing behavior. Among the antecedents of knowledge sharing behavior, codification effort, enjoyment in helping others and community-related outcome expectation are significant, but image and reciprocity are not. The research results help to derive practical strategic implications related to knowledge sharing to activate a virtual community.

• Korean Journal of Culture and Social Issue
• /
• v.18 no.2
• /
• pp.153-167
• /
• 2012

This study was focused on five factors of Hasteful behavior and conducted to examine characteristics of the factors with relationship between Hasteful behavior and type A behavior. 207 adults(18-59 aged) answered the Hasteful Behavior Questionnaire and Type A Behavior Questionnaire. 136 items in established type A behavior or type A personality scale were collected. 136 items were reduced to 6 factors and 29 items. The components of Type A Behavior consists of success striving, impatience, failure anxiety, job immersion, activity, quickness. The factor analysis of Hasteful Behavior resulted in five factors. This was consistent in priority research. Each factors of Hasteful Behavior and type A behavior showed positive correlation. Hasteful behavior under time pressure had positive relations with success striving, failure anxiety, job immersion, activity, and quickness. Hasteful behavior under uncomfortable or isolation had a relation with impatience. Hasteful behavior in bordem had positive relations success striving, activity and quickness. Hasteful behavior in expectation of rewards had positive relations with success striving, impatience, failure anxiety, and quickness.

• (The) Korean Journal of Educational Psychology
• /
• v.31 no.3
• /
• pp.563-587
• /
• 2017

The purpose of this study was to understand the ways to enhance academic help-seeking by analyzing the structural relations among individual(achievement goal orientation) and contextural (teacher behaviors and classroom climate) factors known to affect help-seeking, one of the effective self-regulated learning strategies, for upper elementary students. More specifically, it explored the mediational roles of general classroom climate and student achievement goal orientation in the relation between supportive teacher behaviors and student academic help-seeking. A survey was administered to 315 fifth- or sixth-grade students in three elementary schools and the data from the survey was analyzed. Main results are as follows. First, supportive and learning-oriented teacher behaviors with high expectation related to more cohesive and positive classroom climate and more adaptive achievement goal such as mastery goal. Positive classroom climate played an important role in improving student mastery goal, and only mastery goal among different types of achievement goal orientation had a positive prediction of student help-seeking. Second, teacher behaviors significantly predicted student help-seeking through a double mediation of classroom climate and student mastery goal, which showed that classroom contextual factors and student individual factors interacted for help-seeking. These results suggest that the role of teachers as well as the mastery goal of students are important for enhancing students' help-seeking behavior as an adaptive learning strategy.

• Health Policy and Management
• /
• v.8 no.1
• /
• pp.266-282
• /
• 1998

We model that separation is a rational decision to resolve the inherent uncertainty about quality form the matching process. In health insurance markets, quality of services is revealed to consumers only after enrollment. Discrepancies between the expected and realizeed quality lead utility-maximizing enrollees to disenroll if they find a better alternative. Accordingly, factors that reduce this discrepancy will decrease disenrollment. The firm-level empirical analysis shows that disenrollment is relatively small in markets where the reputation effect works efficiently becuse consumers can predict the expected quality accurately.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.11a
• /
• pp.927-929
• /
• 2021

최근 펫팸족(Pet-Family)과 같이 반려동물을 가족처럼 생각하는 가구가 증가하면서 반려동물 시장이 크게 성장하고 있다. 이러한 이유로 본 논문에서는 반려동물의 객체 식별을 통한 객체 분할과 신체 좌표추정에 기반을 둔 반려동물의 행동 분류 방법을 제안한다. 이 방법은 CCTV를 통해 반려동물 영상 데이터를 수집한다. 수집된 영상 데이터는 반려동물의 인스턴스 분할을 위해 Mask R-CNN(Region Convolutional Neural Networks) 모델을 적용하고, DeepLabCut 모델을 통해 추정된 신체 좌푯값을 도출한다. 이 결과로 도출된 영상 데이터와 추정된 신체 좌표 값은 CNN(Convolutional Neural Networks)-LSTM(Long Short-Term Memory) 모델을 적용하여 행동을 분류한다. 본 모델을 바탕으로 행동을 분석 및 분류하여, 반려동물의 위험 상황과 돌발 행동에 대한 올바른 대처를 제공할 수 있는 기반을 제공할 것이라 기대한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.592-593
• /
• 2022

다중 데이터 기반의 행동 인식 과정에서 데이터 수집 반경이 비교적 제한되는 영상 데이터의 결측에 대한 보완이 요구된다. 본 논문에서는 6축 센서 데이터를 이용하여 결측된 영상 데이터를 생성함으로써 행동 인식의 성능을 개선하는 방법을 제안한다. 가속도와 자이로 센서로부터 수집된 행동 데이터를 이용하여 GAN(Generative Adversarial Network)을 통해 영상에서의 관절(Skeleton) 움직임에 대한 데이터를 생성하고자 한다. 이를 위해 DeepLabCut 기반 모델 학습을 통해 관절 좌표를 추출하며, 전처리된 센서 시퀀스 데이터를 가지고 GRU 기반 GAN 모델을 통해 관절 좌표에 대한 영상 시퀀스 데이터를 생성한다. 생성된 영상 시퀀스 데이터는 영상 데이터의 결측이 발생했을 때 대신 행동 인식 모델의 입력값으로 활용될 수 있어 성능 향상을 기대할 수 있다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.655-656
• /
• 2023

반려동물 웨어러블 시장의 성장함에 따라 반려동물의 행동 패턴을 측정하고 분석할 수 있는 센서데이터가 활용되고 있다. 본 논문에서는 반려동물 수면 패턴 모니터링을 위한 행동 분류 모델을 제안한다. 6축 센서 데이터를 활용한 가속도 및 자이로센서 데이터를 입력 데이터로 사용한다. 제안된 모델은 ResNet을 통해 시간에 따라 가속도 및 자이로센서 데이터의 특징을 추출한 후 LSTM을 사용하여 시계열 정보를 고려한 행동 분류를 수행한다. 이러한 과정을 통해 정확한 행동 패턴 분석이 가능하게 되며 반려동물의 건강 관리 및 수면 질 개선에 기여할 것으로 기대한다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.5-6
• /
• 2023

본 연구에서는 언텍트 소비가 일반화됨에 따라 소상공인들을 지원하기 위해 캡티브-포털을 활용하여 주문하는 등의 시스템을 구축하고 있으며, 이에 상권 내 방문자들의 주문 정보를 기반으로 개인의 선호나 취향을 고려하고 기존 방문 순서를 고려하여 다음 방문지를 추천할 수 있는 모델을 개발하고자 한다. 모델 개발을 위한 데이터셋으로는 캡티브-포털을 통해 수집되는 변수 항목과 유사한 위치기반 SNS 데이터인 Foursquare 데이터를 활용했다. 본 논문에서는 데이터셋의 변수 중 상호명을 기반으로 22개의 행동 유형 카테고리로 묶어 현재 행동 유형 이후에 다음에 이어질 행동 유형을 예측하는 것을 제안한다. 개인 별 세션 기반의 데이터셋을 LightMove 알고리즘을 활용하여 행동유형 예측을 임베딩 차원의 변경하여 실험한 결과 500차원에서 Top-5가 82.72의 성능을 보임을 확인했다. 향후 국내 상권에 맞는 방문지 추천 시스템이 개발된다면 방문지 추천을 활용하여 다양한 마케팅 전략을 수립이 가능해질 수 있고, 이를 통해 지역 상권이 활성화될 것으로 기대된다.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.12
• /
• pp.230-241
• /
• 2017

The main purpose of this study was to empirically investigate the effects of participation behavior and performance improvement on motivation factors of Korean university students which participated in LINC by utilizing Vroom's Expectancy Theory. Three motivation factors of valence, instrumentality, and expectancy were examined in this study. In addition, two different models (valence and force model) analyzed the causal relationships regarding participation behavior and performance improvement. 236 data were collected and findings of this study were as follows: First, comparative analysis between demographic characteristics including university, major, and residence had no significant differences in mean value. However, females had higher levels of recognition related to valence (attractiveness) relative to males. Second, valence and the force model were significant predictors of LINC participation behavior and performance improvement. Furthermore, the coefficient of determination and beta coefficient of the force model were higher compared with the valence model. Third, the level of mediation effects including direct, indirect, and total effect of the force model was higher than the valence model. LINC participation behavior had a partial mediating effect between the three motivation factors and performance improvement variable.