• Communications for Statistical Applications and Methods
• /
• v.21 no.6
• /
• pp.501-512
• /
• 2014

Stylized facts on asset return are fat-tail, asymmetry, volatility clustering and structure changes. This paper simultaneously captures these characteristics by introducing a multi-regime models: Finite mixture distribution and regime switching GARCH model. Analyzing the daily KOSPI return from $4^{th}$ January 2000 to $30^{th}$ June 2014, we find that a two-component mixture of t distribution is a good candidate to describe the shape of the KOSPI return from unconditional and conditional perspectives. Empirical results suggest that the equality assumption on the shape parameter of t distribution yields better discrimination of heterogeneity component in return data. We report the strong regime-dependent characteristics in volatility dynamics with high persistence and asymmetry by employing a regime switching GJR-GARCH model with t innovation model. Compared to two sub-samples, Pre-Crisis (January 2003 ~ December 2007) and Post-Crisis (January 2010 ~ June 2014), we find that the degree of persistence in the Pre-Crisis is higher than in the Post-Crisis along with a strong asymmetry in the low-volatility (high-volatility) regime during the Pre-Crisis (Post-Crisis).

• The Journal of Information Technology
• /
• v.5 no.2
• /
• pp.17-26
• /
• 2002

On this paper, we propose SOSL(Self-Organized Supervised Learning) and it's architecture SOSL is hybrid type neural network. It consists of several CBP (Component Back Propagation) neural networks, and a modified PCA neural networks. CBP neural networks perform supervised learning procedure in parallel to clustered and complex input patterns. Modified PCA networks perform it's learning in order to transform dimensions of original input patterns to lower dimensions by clustering and local projection. Proposed SOSL can effectively apply to neural network learning with large input patterns results in huge networks size.

• Journal of the Semiconductor & Display Technology
• /
• v.16 no.4
• /
• pp.30-35
• /
• 2017

Moving object segmentation from a nonstationary camera is a difficult problem due to the motion of both camera and the object. In this paper, we propose a new confidence-based background subtraction technique from moving camera. The method is based on clustering of motion vectors and generating adaptive multi-homography from a pair of adjacent video frames. The main innovation concerns the use of confidence images for each foreground and background motion groups. Experimental results revealed that our confidence-based approach robustly detect moving targets in sequences taken by a freely moving camera.

• Electronics and Telecommunications Trends
• /
• v.37 no.4
• /
• pp.1-10
• /
• 2022

As the response to climate change becomes a more pressing global issue, so do expectations for climate change in the green information and communication technology (ICT) industry and the possibility of solving environmental problems through ICT. However, because the green ICT industry is still in its early stages, there is little research on it. Understanding the startup ecosystem in the industry is helpful for recognizing innovation trends in emerging technologies such as green ICT. In this regard, this paper investigates the current state and characteristics of the green ICT ecosystem and presents implications based on an examination of startup venture capital investment trends and submarket identification in the green ICT industry as emphasized by the carbon neutrality paradigm shift. The analysis included 4,807 companies and 3,990 funding records, as well as exploratory data analysis and "k-means" clustering techniques.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.9
• /
• pp.13-25
• /
• 2021

In the age of digital innovation based on the Internet, Information and Communication and Artificial Intelligence technologies, huge amounts of datasets are being generated, collected, accumulated, and opened on the web by various public institutions providing useful and public information. In order to analyse, gain useful insights and information from data, Formal Concept Analysis(FCA) has been successfully used for analyzing, classifying, clustering and visualizing data based on the binary relation between objects and attributes in the dataset. In this paper, we present an approach for enhancing the analysis of relational attributes of data within the extended framework of FCA, which is designed to classify, conceptualize and visualize sets of objects described not only by attributes but also by relations between these objects. By using the proposed tool, RCA wizard, several experiments carried out on some public open datasets demonstrate the validity and usability of our approach on generating and visualizing conceptual hierarchies for extracting more useful knowledge from datasets. The proposed approach can be used as an useful tool for effective data analysis, classifying, clustering, visualization and exploration.

• Journal of Technology Innovation
• /
• v.23 no.4
• /
• pp.151-175
• /
• 2015

The general public is a key stakeholder in the science and technology domain. However, traditional approaches require substantial efforts and resources to analyze how does the general public understand science and technology issues. We applied the topic modeling, a form of text clustering, to the texts about the nuclear power which were posted on an online space in order to explore the general public's thoughts on the issue. This study investigates the extent to which macro-level events influence understandings of the general public on the science and technology issues and weather these changes in understandings are sustained over time. It examines the possibility of applying topic modeling in narrowing a perception gap between the general public and the experts through a near-real-time monitoring of the public interests and perceptions about the science and technology issues.

• Proceedings of the Technology Innovation Conference
• /
• 2005.06a
• /
• pp.69-91
• /
• 2005

나노 기술은 국가와 기업 차원에서 전략적으로 활발하게 연구, 개발되고 있는 분야로 기술과 산업 전 분야에 걸쳐 있으며 계속 확대되고 있다. 이중 탄소나노튜브는 약 10년여 전부터 활발히 연구되고 있으며, 소재, 소자 등 나노 기술의 전 분야에 걸쳐서 적용될 수 있는 물질로 현재까지 실용화된 기술이나 제품이 많이 나오지는 않았으나 앞으로 다양한 적용분야가 창출될 전망이다. 그러나 전략적인 기술 개발을 위한 연구는 미비한 실정이다. 전략적이고 효과적인 기술 개발의 대안으로 특허 분석(patent analysis)이 제시되고 있으며 나노 기술에 대한 전반적인 특허 분석은 각 기업 연구소나 국가 연구소에서 진행되어 왔으나, 여기서 도출된 정보만으로는 전략적인 문제들을 분석하기에는 미흡하였다. 따라서 전략적인 기술 개발을 위한 특허 분석의 필요성이 제기되고 있다. 이에 본 연구에서는 특허의 인용(citation) 정보를 이용하여 특허와 특허 사이의 관계를 분석하고 이를 바탕으로 관련된 특허를 군집화(clustering) 하여 각 군집의 특성을 살펴보았다. 먼저 탄소나노튜브에 관한 특허를 USPTO 자료를 통해 추출하고, 이 특허들이 인용한 특허들을 추출하여 각 특허들이 어느 정도로 관련이 되어 있는가를 상관 행렬(correlation matrix)을 이용하여 조사하였다. 인용 특허의 분석은 기초 기술 분야에 이르기까지 광범위하게 조사하여 기술별로 특허를 군집화 하였을 뿐만 아니라, 특허 출원인 정보도 활용하여 군집화 하였다. 이를 통하여 기술별, 특허 출원인 별로 특허의 군집을 만들어서 각 군집의 특성을 도출하였고, 마지막으로 각 군집의 특성에 맞게 기술 개발측면에서 전략적인 시사점을 도출하였다. 각각의 군집들은 시간이 지날수록 확대, 통합되는 결과를 보이고 있으며, 융합기술화 되고 있다. 군집을 이루는 주요한 특성으로는 출원인 정보가 큰 비중을 차지하였으며, 이는 출원인의 기술 개발 전략과 이에 따른 기술개발경로나 특허 전략과 관련이 있는 것으로 나타났다. 다양한 기술에 관하여 군집의 크기가 계속 확대되고 있는 경향을 보이고 있기 때문에, 앞으로 지속적인 연구를 통하여 향후 특허 관리와 특허 전략에 적용할 수 있도록 하는 것이 중요하다.

• Journal of Korea Technology Innovation Society
• /
• v.14 no.spc
• /
• pp.1187-1208
• /
• 2011

This research aims to propose a systematic approach to identify emerging technology-product portfolio for small and medium enterprises (SMEs). Firstly, operational definition of emerging technology for SMEs is presented. Secondly, research framework is suggested and case study to show usefulness of the newly proposed framwork is analyzed. In detail, reference patent set which represent company's capabilities and business area are constructed. The research constructs patent data set for bibliometric analysis using reference patent set and citing patents to 2nd level. Clustering (expert judgement) and keyword based bibliometric approach are used. Then, cluster activity index (AI) and relevance index (RI) comparing with reference patent set are estimated. With emerging technology-product portfolio using AI and RI, a firm can identify emerging technology-product area and monitoring area.

• The Journal of the Korea Contents Association
• /
• v.21 no.11
• /
• pp.64-76
• /
• 2021

Culture technology (CT) becomes important in the recent environment where digital technology drives content-based innovations. However, technological trends of CT have not been systematically discussed. Especially, the trends of CT should be analyzed from the national perspective, because CT has grown with the help of government-driven innovation. Therefore, this paper aims to analyze CT trends focusing on national research projects. We collected data on CT from the national science and technology information service (NTIS) database, analyzed the keyword co-occurrence network, and identified the patterns of technological innovation using a clustering analysis. As a result, we found that CT has contributed to the digital content and cultural media, and has been actively developed with the help of machine learning technique. Especially, due to the rise of Covid-19, the non-face-to-face online content is rapidly increasing. This study provides important clues for understanding, analyzing CT trends.

• Journal of Science and Technology Studies
• /
• v.9 no.1
• /
• pp.33-55
• /
• 2009

This article analyzes the evolutionary process of Daedeok Science Town from the standpoint of the life of science and technology poles. It deals the theoretical discussion on the science and technology poles, and investigates the history of Daedeok Science Town dividing into the period of conception, construction, and clustering. Daedeok Science Town has been developed from research and academic city through special research parks to innovative clusters, and nowadays emphasizes networking, interaction, and commercialization. This study shows the types of science and technology poles can be analyzed by historical consideration of specific Korean case.