• 프린팅코리아
• /
• v.11 no.4
• /
• pp.56-69
• /
• 2012

환경이슈는 점점 더 적극적으로 대처해야 할 과제로 부각되고 있다. 인쇄업계도 이러한 변화에 능동적으로 대응하기 위해 환경이슈와 관련한 개발동향, 글로벌 트렌드, 성공사례 등에 대한 최신정보의 습득이 필요한 실정이다. 이에 본지는 친환경 특별기획을 마련, 그 첫번째 순서로 친환경인쇄, '잉크가 시작이다'를 준비했다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.3
• /
• pp.614-621
• /
• 2006

We have implemented a multiclassifier learning approach in a GA-based inductive learning environment that learns classification rules that are similar to rules used in PROSPECTOR. In the multiclassifier learning approach, a classification system is constructed with several classifiers that are obtained by running a GA-based learning system several times to improve the overall performance of a classification system. To implement the multiclassifier learning approach, we need a decision-making scheme that can draw a decision using multiple classifiers. In this paper, we introduce two decision-making schemes: one is based on combining posterior odds given by classifiers to each class and the other one is a voting scheme based on ranking assigned to each class by classifiers. We also present empirical results that evaluate the effect of the multiclassifier learning approach on the GA-based inductive teaming environment.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.759-762
• /
• 2018

최근 들어 점차 지능형 서비스 로봇들이 인간의 실생활 속으로 들어옴에 따라, 로봇 스스로 다양한 물체들을 효과적으로 조작할 수 있는 지식을 습득하는 기계 학습 기술들이 매우 주목을 받고 있다. 전통적으로 로봇 행위 학습 분야에는 강화 학습 혹은 심층 강화 학습 기술들이 주로 많이 적용되어 왔으나, 이들은 대부분 물체 조작 작업과 같이 다차원 연속 상태 공간과 행동 공간에서 최적의 행동 정책을 학습하는데 여러가지 한계점을 가지고 있다. 따라서 본 논문에서는 전문가의 데모 데이터를 활용해 보다 효율적으로 물체 조작 행위들을 학습할 수 있는 모방 학습과 강화 학습의 통합 프레임워크를 제안한다. 이 통합 프레임워크는 학습의 효율성을 향상시키기 위해, 기존의 GAIL 학습 체계를 토대로 PPO 기반 강화 학습 단계의 도입, 보상 함수의 확장, 상태 유사도 기반 데모 선택 전략의 채용 등을 새롭게 시도한 것이다. 다양한 성능 비교 실험들을 통해, 본 논문에서 제안한 통합 학습 프레임워크인 PGAIL의 우수성을 확인할 수 있었다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.13 no.2
• /
• pp.56-68
• /
• 2008

Technology adoption has been an important issue for researchers and practitioners. In this paper, we identified the relationships between intention of technology adoption and technological affinity and other psychometric characteristics. We believe that technology affinity may be one of the general psychometric traits of individuals, and thus people have a different affinity levels which may influence the technology adoption intention. As a result it was found that need for cognition and self efficacy had positive influence on technology affinity, and the affinity also positively affected adoption intention. In addition, it was also found that technology affinity displayed a mediating role for the consumers adoption intention with need for cognition and self efficacy.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.4
• /
• pp.299-314
• /
• 2017

Technological advances in media have expanded users' consciousness. At the same time, users have changed from passive into active voice by interacting media. The emergence of mobile made different structures and contents compared to the past. Especially, Korean culture of mobile converted original media channels to contents in a category. Plus, the usage structure of internet of this time converges in massive portal sites. It is because that the structure has aspect of emitting through remediation in the sites. Also, Korean massive portal sites have provided specific service named 'real-time issues'. This is not only the unique way of offering information that exists in Korea but also high usability of getting issues. We therefore considered the meaning of durability of real-time issues in the view of journalism, compared original media channels. Then, this paper identified the user attributes for acquisition of information following ways using informal and formal data from Korean massive portal sites named 'Daum' and 'Naver'.

• Journal of Korean Society of Forest Science
• /
• v.113 no.1
• /
• pp.1-13
• /
• 2024

This study used the results of the 2022 national survey of forest welfare facilities and space users conducted in the Republic of Korea to identify the associations and combination patterns among use of forest welfare facilities and various user characteristics. In addition, a correspondence analysis-namely, a categorical data analysis-was performed regarding four facilities: National Natural Recreation Forest, National Soopchewon, National Forest Trail, and National Healing Forest. The cross-analysis revealed highly significant differences (p < .001) in the age, household income per month, means of transportation, companion type, path to information acquisition, and purpose of visit among the users of forest welfare facilities. Moreover, the results of the correspondence analysis demonstrated that users of National Soopchewon and National Healing Forest were distinguished from users of the other forest welfare facilities based on demographic characteristics such as age and household income per month. In addition, the means of transportation of National Forest Trail users differed from that of users of the other forest welfare facilities. Moreover, users of National Soopchewon were associated with the type of public group visit, acquisition of information from the websites of related institutions and facilities and via online portals, and the purpose of participating in education and experiencing education programs. Therefore, the companion type, path to information acquisition, and purpose of visit on the part of users of National Soopchewon could be differentiated from those of users of other forest welfare facilities. The findings of this study revealed the associations and combinations among forest welfare facilities and the characteristics of users of such facilities, with the scope of utilization of the forest welfare facility user data being expanded via correspondence analysis.

• Journal of KIISE:Software and Applications
• /
• v.34 no.9
• /
• pp.804-816
• /
• 2007

Sample training data for machine learning often contain irrelevant information or redundant concept. It is also the case that the original data may include noise. If the information collected for constructing learning model is not reliable, it is difficult to obtain accurate information. So the system attempts to find relations or regulations between features and categories in the teaming phase. The feature selection is to remove irrelevant or redundant information before constructing teaming model. for improving its performance. Existing feature selection methods assume that the distribution of documents is balanced in terms of the number of documents for each class and the length of each document. In practice, however, it is difficult not only to prepare a set of documents with almost equal length, but also to define a number of classes with fixed number of document elements. In this paper, we propose a new feature selection method that considers the impurities among the words and unbalanced distribution of documents in categories. We could obtain feature candidates using the word impurity and eventually select the features through unbalanced distribution of documents. We demonstrate that our method performs better than other existing methods via some experiments.

• Korean Journal of Construction Engineering and Management
• /
• v.22 no.3
• /
• pp.40-51
• /
• 2021

MLS (Mobile Laser Scanning) which is a scanning method done by moving the LiDAR (Light Detection and Ranging) is widely employed to capture indoor PCD (Point Cloud Data) for floor plan generation in the AEC (Architecture, Engineering, and Construction) industry. The movement and rotation of LiDAR in the scanning phase cause deformation (i.e. skew) of PCD and impose a significant impact on quality of output. Thus, a de-skewing method is required to increase the accuracy of geometric representation. De-skewing methods which use position and pose information of LiDAR collected by IMU (Inertial Measurement Unit) have been mainly developed to refine the PCD. However, the existing methods have limitations on de-skewing PCD without IMU. In this study, a novel algorithm for de-skewing 2D PCD captured from MLS without IMU is presented. The algorithm de-skews PCD using scan matching between points captured from adjacent scan positions. Based on the comparison of the deskewed floor plan with the benchmark derived from TLS (Terrestrial Laser Scanning), the performance of proposed algorithm is verified by reducing the average mismatched area 49.82%. The result of this study shows that the accurate floor plan is generated by the de-skewing algorithm without IMU.

• 한국경영정보학회:학술대회논문집
• /
• 2008.06a
• /
• pp.86-90
• /
• 2008

본 연구는 개인이 이른바 '유비쿼터스 개인지식관리 시스템'을 갖추고 있을 때 발생하게 될 것으로 예상되는 마케팅 메시지로서의 지식의 역할에 대한 개념적 연구를 수행한다. 기존의 마케팅 광고 메시지는 인간의 생물학적인 감각과 인지, 기억 능력에 의존하였기 때문에 반복적이고 자극적이어야만 했다. 그러나 유비쿼터스 환경에서는 UPA(Ubiquitous Personal Assistant)를 활용하여 개인이 소비하는 컨텐트를 이음매 없이(Seamless) 습득, 저장할 수 있게 되며, 이러한 변화는 마케팅과 광고의 방법에도 변화를 가져오게 될 것으로 판단된다. 이에 본 연구에서는 유비쿼터스 환경에서의 개인지식관리를 통하여 변화되는 마케팅과 광고의 모습을 나타내고자 하며, 이를 위해 시나리오를 제시하고 분석한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04a
• /
• pp.349-351
• /
• 2002

한국현재의 컴퓨팅환경은 Text위주의 Command Line상에서의 입출력에서 GUI(Graphic User Interface)환경으로 전환되었다. 이는 사용자에게 좀더 친근한 방법으로의 컴퓨팅환경을 제공하고 있는 것이다. 하지만 아직까지 그러한 환경에 익숙해지기 위해서는 많은 습득시간이 필요하며 또한, 응용프로그램 간의 인터페이싱 기능 등을 익히기 위해서는 추가적인 학습을 통해야 원활한 작업을 수행할 수 있다. 이를 해결하고자 본 연구는 음성인식/ 합성과, 현재 음성마크업 언어인 VoiceXML 등을 통해서 모색해보고자 한다.