• Journal of KIISE
• /
• v.43 no.10
• /
• pp.1124-1130
• /
• 2016

A controlled group is closed compared to other organizations, which hinders collection of data and accurate analysis, so that it is hard to evaluate a controlled group's power structure and predict future changes using usual analytical methods including sociological approach. Analyzing a controlled group using SNA can allow for evaluation of inner power structure by revealing the relationships between members and identifying members with central roles given limited data. In this study, in order to evaluate changes in power structure, time-sequential SNA research was conducted by analyzing eigenvector centrality, which reflects individual influence and reveals the overall power structure. The result showed an improvement in accuracy compared to other centralities that contain individual degree or closeness, and made it possible to presume structural changes such as promotion or purge of a member.

• Journal of KIISE
• /
• v.44 no.7
• /
• pp.674-679
• /
• 2017

In big-data environments wherein large amounts of text documents are produced daily, titles are very important clues that enable a prompt catching of the key ideas in documents; however, titles are absent for numerous document types such as blog articles and social-media messages. In this paper, a title-generation model for which sequence-to-sequence RNNs with attention and copying mechanisms are employed is proposed. For the proposed model, input sentences are encoded based on bi-directional GRU (gated recurrent unit) networks, and the title words are generated through a decoding of the encoded sentences with keywords that are automatically selected from the input sentences. Regarding the experiments with 93631 training-data documents and 500 test-data documents, the attention-mechanism performances are more effective (ROUGE-1: 0.1935, ROUGE-2: 0.0364, ROUGE-L: 0.1555) than those of the copying mechanism; in addition, the qualitative-evaluation radiative performance of the former is higher.

• Korean Journal of Cognitive Science
• /
• v.13 no.4
• /
• pp.54-54
• /
• 2002

This study was conducted to explore whether an automatic evaluation effect (i.e.,congruency effect and positive priming effect) of emotional properties which have been found in the racial prejudice study can be applied to the gender properties. Experiment 1 employing a short SOA (150ms-250ms) naming task showed a priming effect was larger in the positive prime-positive target condition than negative prime-negative target condition but not congruent effect. Experiment 2 employing a long SOA (500ms-1000ms) naming task didn't yield either positive priming effect or congruency effect. The congruency effect and the positive priming effect which is ubiquitous phenomenon in the generic cognitive concepts network were not found in the gender properties. Therefore, it was suggested that the social information including prejudice or stereotype can be differently processed according to the value of emotionality.

• Korean Journal of Cognitive Science
• /
• v.13 no.4
• /
• pp.55-65
• /
• 2002

This study was conducted to explore whether an automatic evaluation effect (i.e.,congruency effect and positive priming effect) of emotional properties which have been found in the racial prejudice study can be applied to the gender properties. Experiment 1 employing a short SOA (150ms­250ms) naming task showed a priming effect was larger in the positive prime-positive target condition than negative prime-negative target condition but not congruent effect. Experiment 2 employing a long SOA (500ms­1000ms) naming task didn't yield either positive priming effect or congruency effect. The congruency effect and the positive priming effect which is ubiquitous phenomenon in the generic cognitive concepts network were not found in the gender properties. Therefore, it was suggested that the social information including prejudice or stereotype can be differently processed according to the value of emotionality.

• Archives of design research
• /
• v.14 no.1
• /
• pp.7-16
• /
• 2001

Human beings have been leading their life and desirable social life through communication. Human communication has been changed and expanded through language, letters, printing media, broadcasting media, and more recently network communication media. Since the invention of characters, books have been the human beings' communication medium that have the longest history, and they have preserved and succeeded human spiritual world. Children's picture books are also a communication medium composed of the transmitter of the information, message, medium, and the receiver, and makes the process of communication. The sender: writer, illustrator, and editor, analyzes children, special receiver, symbolizes messages, make up contents and make communication come true by selecting media and effectively delivering them to children. Considering the developmental characteristics of the receiver, children and their desires, visual media such as the writings and illustrations. The writings and illustrations are the most appropriate for communication with children. The first, picture books are for the communication between grown-ups and children, and also that with society for helping children find their identity and perform their roles. The second, through the message of writings and illustrations young children acquire human communication ability of this age. That is, they experience and learn visual communication and letter communication

• Journal of KIISE
• /
• v.43 no.8
• /
• pp.910-918
• /
• 2016

A tensor is a multi-dimensional array that represents many data such as (user, user, time) in the social network system. A tensor generator is an important tool for multi-dimensional data mining research with various applications including simulation, multi-dimensional data modeling/understanding, and sampling/extrapolation. However, existing tensor generators cannot generate sparse tensors like real-world tensors that obey power law. In addition, they have limitations such as tensor sizes that can be processed and additional time required to upload generated tensor to distributed systems for further analysis. In this study, we propose TeT, a distributed tera-scale tensor generator to solve these problems. TeT generates sparse random tensor as well as sparse R-MAT and Kronecker tensor without any limitation on tensor sizes. In addition, a TeT-generated tensor is immediately ready for further tensor analysis on the same distributed system. The careful design of TeT facilitates nearly linear scalability on the number of machines.

• Journal of Internet Computing and Services
• /
• v.18 no.6
• /
• pp.93-100
• /
• 2017

In multi-media, multi-channel era, video viewing using various digital devices via the Internet, not the existing broadcasting network, is emerging as a new broadcasting viewing behavior. OTT service is regarded as the main service of this change of viewing behavior. It is a video service that can seamlessly use any content desired by the user at any time, at any time, any device, any contents, if the Internet connection is available. The purpose of this study is to investigate the factors affecting the satisfaction of OTT service by applying the technology acceptance model (TAM). As a result of analyzing through 303 questionnaires of the early users of OTT service, social pressure, perceived popularity, perceived cost, user reputation, individual innovation, and esthetics set as external factors in this study are partially affects perceived usefulness, perceived ease of use, and perceived enjoyment. In addition, perceived usefulness, ease of use, and enjoyment are directly influencing satisfaction. Based on these results, we discuss the theoretical and practical implications and propose future research direction.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2008.10b
• /
• pp.218-234
• /
• 2008

중국은 2008년 현재 메신저 이용자 수와 블로그 등의 웹 2.0 이용자 수에 있어서 세계 1위이다. 그리고 중국의 2007년 850억 달러의 인터넷 디지털 시장은 2015년에 2640억 달러로 성장할 전망이다. 이와 같이 인터넷 시장의 규모와 성장 가능성으로 인해 중국 인터넷 비즈니스의 중요성이 높아지고 있다. 이러한 필요성에 따라서 본 연구에서는 중국의 인터넷 동향 중에서 가장 큰 주목을 끌고 있는 웹 2.0 기반의 사회 연결망 서비스 분야에서 다음 세 가지의 연구 질문을 설정하고 그 답을 찾아보았다. 첫째, 중국 사회 연결망 서비스의 주요 특징은 무엇인가? 이러한 질문에 답하기 위해 본 연구에서는 문헌연구를 바탕으로 중국의 대표적인 사회 연결망 서비스 업체로 Tencent QQ와 Sina Poco를 살펴보았고, 중국 사회 연결망 서비스 이용자의 가입 이유, 갱신 이유, 주요 내용, 유상 서비스에 대한 태도 등을 살펴보았으며, 중국 인터넷 사용자의 집단별 특징에 대해 살펴보았다. 둘째, 중국 사회 연결망 서비스의 이용에 영향을 미치는 주요 요인들은 무엇인가? 이러한 질문에 답하기 위해 동기부여이론, TAM 이론. 관련 선행연구 등을 검토하여 중국 사회 연결망 서비스의 이용에 영향을 미치는 요인들로서 사용자 참여, 사회적 영향, 네트워크 효과, 유용성, 시스템 품질 등을 도출하였고 PLS를 이용한 데이터 분석을 통해 검증하였다. 검증결과, 사회 연결망 서비스 이용에 유용성이 가장 큰 영향을 미치고, 다음으로 네트워크 효과, 사용자 참여, 시스템 품질 순으로 유의한 영향을 미치는 것으로 나타났다. 반면, 사회적 영향은 유의하지 않은 것으로 나타났다. 셋째, 중국과 한국은 어떠한 점에서 차이가 있는가? 가장 두드러진 차이는 사회적 영향에 있어서 한국은 유의 적인데 비해 중국이 유의적이지 않다는 점이다. 이는 한국과 중국 사이에 국가 문화적 차이가 존재하기 때문에 발생했다고 생각할 수도 있고, 중국이 인터넷 성장기에 있는데 비해 한국은 인터넷 성숙기에 있는 등의 기술적, 사회적, 경제적 발달 과정상의 차이에 의한 것이라고도 볼 수 있다.

• Journal of Digital Convergence
• /
• v.14 no.11
• /
• pp.481-488
• /
• 2016

This study is about comparing the mobile user experience of Facebook and Instagram which are most often used among the recent SNSs by the people in their 30s and under. This study analyzed the user experience level after dividing the user experience factors through the Creating Pleasurable Interfaces model, and suggested the mean analysis as well as the result of Wilcoxon rank test which is a nonparametric statistics method. As a result of study, the Display information visually factor in functional factor and the configuration of the main page in convenient factor were a statistically significant difference in the mobile user experience of Facebook and Instagram. It is expected that this study may help seeking the user experience factors to be promoted preferentially in a competitive situation through the statistical comparative evaluation of the experience of two SNS users.

• Korean Institute of Interior Design Journal
• /
• v.15 no.6 s.59
• /
• pp.131-138
• /
• 2006

Contemporary society is lately going through various changes in social, economic, and cultural aspects due to rapid growth of information and computer technology. The digital revolution generated from the development of network reflects the decentralizing characteristic that connects local and personal interconnections. The following paper is designed to develop a basic understanding of digital architecture and show that digital virtual reality is not accidental phenomenon but it is actually evolved from nature and recreated in cyber space. The relationship between nature and digital architecture will be explained through the background research of how digital architecture was born and how it has been transformed. Also, it is designed to find out modern trends of digital architecture through diagrams and virtual space of hyper architecture and follow up the new phenomena appearing in the field of architecture. It is said that digital architecture is a new trend of architecture created from computer bit but all architecture that has form and shape cannot be independent of nature. Nature is the root of everything and even mechanical abstraction such as 'digital' can find its form in the immanence melted in the pure essence of nature. The research found that the nature approach of digital architecture also needs no more than 5 sequential references to find its genuine sketch as the hypertext theory shows we need only 5 sequential references to prove we are all related. Additionally, the cyber space that is become a general living space and a indispensable factor of digital architecture is a space that has obscurity and more open culture. It represents the characteristics of contemporaries in various aspects of society.