• Journal of Information Technology Applications and Management
• /
• v.23 no.4
• /
• pp.25-39
• /
• 2016

Asset prices decline sharply and stock markets collapse when financial crisis happens. Recently we have encountered more frequent financial crises than ever. 1998 currency crisis and 2008 global financial crisis triggered academic researches on early warning systems that aim to detect the symptom of financial crisis in advance. This study proposes a risk recovery index for detection of good opportunities from financial market instability. We use SVM classifier algorithms to separate recovery period from unstable financial market data. Input variables are KOSPI index and V-KOSPI200 index. Our SVM algorithms show highly accurate forecasting results on testing data as well as training data. Risk recovery index is derived from our SVM-trained outputs. We develop a trading system that utilizes the suggested risk recovery index. The trading result records very high profit, that is, its annual return runs to 121%.

• Journal of the Korea Society for Simulation
• /
• v.6 no.2
• /
• pp.45-57
• /
• 1997

MAC (media access control) bridge is used to interconnect separate LANs and to relay frames between the BLANs (bridged LANs). Bridge architecture consists of MAC entity, MAC relay entity and bridge protocol entity protocol entity and performs learning, filtering and forwarding functions using filtering database. In this paper, we simulate these functions of bridge and the STP (spanning tree protocol). The STP derives an active topology from an arbitrarily connected BLAN. Our simulation model assumes a BLAN consisted of three bridge forming a closed loop. In order to remove the loop, each bridge process exchanges configruation BPDU (bridge protocol data unit0 with other bridge processes connected to the bridge itself. To simulate the communication between bridges, we implement the IPC (inter-process communication) server using message queues. Our simulation results show that the assumed BLAN contains no closed loop and then there is no alternative route and no unnecessary traffic.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.9-14
• /
• 2002

We propose a new algorithm for blind source separation (BSS), in which frequency-domain independent component analysis (FDICA) and time-domain ICA (TDICA) are combined to achieve a superior source-separation performance under reverberant conditions. Generally speaking, conventional TDICA fails to separate source signals under heavily reverberant conditions because of the low convergence in the iterative learning of the inverse of the mixing system. On the other hand, the separation performance of conventional FDICA also degrades significantly because the independence assumption of narrow-band signals collapses when the number of subbands increases. In the proposed method, the separated signals of FDICA are regarded as the input signals for TDICA, and we can remove the residual crosstalk components of FDICA by using TDICA. The experimental results obtained under the reverberant condition reveal that the separation performance of the proposed method is superior to that of conventional ICA-based BSS methods.

• Journal of KIEE
• /
• v.11 no.1
• /
• pp.1-7
• /
• 2001

Various methods for blind source separation (BSS) are based on independent component analysis (ICA) which can be viewed as a nonlinear extension of principal component analysis (PCA). Most existing ICA methods require certain nonlinear functions (which leads to higher-order statistics) depending on the probability distributions of sources, whereas PCA is a linear learning method based on second-order statistics. In this paper we show that the PCA can be applied to the task of BBS, provided that source are spatially uncorrelated but temporally correlated. Since the resulting method is based on only second-order statistics, it avoids the nonlinear function and is able to separate mixtures of several colored Gaussian sources, in contrast to the conventional ICA methods.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06b
• /
• pp.160-162
• /
• 2006

BP 알고리즘은 지역 최소점이나 고원 문제와 같은 수렴 실패문제와 학습 속도가 느리다고 알려져 있다. 이제까지 알려진 BP 알고리즘의 대체 방법들은 수렴 속도와 인자에 따른 수렴의 안정성에 대한 불균형을 해소하는데 치중했다. 기존의 전통적인 BP 알고리즘에서 발생하는 위와 같은 문제를 해결하기 위하여, 본 논문에서는 적은 용량의 저장 공간만을 요구하며 수렴이 빠르고 상대적으로 안정성이 보장되는 알고리즘을 제안한다. 이 방법은 상위연결(upper connections), 은닉층-출력층(hidden to output), 하위연결(lower connections), 입력층-은닉층(input to hidden)에 대해 개별적으로 훈련을 시키는 분리 학습방법을 적용한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.10a
• /
• pp.1002-1003
• /
• 2019

Face recognition is an extensively researched area of computer vision. Visible, infrared, thermal, and 3D modalities have been used against various challenges of face recognition such as illumination, pose, expression, partial information, and disguise. In this paper we present a multi-modal approach to face recognition using convolutional neural networks. We use visible and thermal face images as two separate inputs to a multi-input deep learning network for face recognition. The experiments are performed on IRIS visible and thermal face database and high face verification rates are achieved.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.22.1-22
• /
• 2001

In this paper we present Multiple-Channel Active Noise Control[ANC] system by employing Independent Component Analysis[ICA] and Adaptive Network Fuzzy Inference System[ANFIS]. ICA is widely used in signal processing and communication and it use prewhiting and appropriate choice of non-linearities, ICA can separate mixed signal. ANFIS controller is trained with the hybrid learning algorithm to optimize its parameters for adaptively canceling noise. This new method which minimizes a statistical dependency of mutual information(MI) in mixed low frequency noise signal and there is no need to secondary path modeling. The proposed implementations achieve more powerful and stable noise reduction than Filtered-X LMS algorithms which is needed for LTI assumption and precise secondary error

• International conference on construction engineering and project management
• /
• 2009.05a
• /
• pp.970-977
• /
• 2009

Organizational personnel must have well-developed interpersonal skills to deal with the different stakeholders and departments, to work at different levels in the hierarchy, and to meet varying performance requirements. Many leadership development and mentoring programs are designed to expose students as well as construction professionals to contemporary leadership techniques and skills. Leadership skills generally separate into three decision-making styles with varying degrees: 1) Autocratic, 2) Participate, and 3) Free-rein. This paper describes the study of leadership styles among 174 construction professionals and addresses the most appropriate leadership style for a project executive and a project manager in relation to compare with the characteristic leadership style and job functions. The study supports the growing importance of leadership skills as a component of managerial functions and provides a benchmark to identify a dominant leadership skill for a specific managerial position.

• Journal of the Korean Society for information Management
• /
• v.40 no.2
• /
• pp.115-135
• /
• 2023

The purpose of this study is to assess the effectiveness of using deep learning language models to extract references automatically and create a reference database for research reports in an efficient manner. Unlike academic journals, research reports present difficulties in automatically extracting references due to variations in formatting across institutions. In this study, we addressed this issue by introducing the task of separating references from non-reference phrases, in addition to the commonly used metadata extraction task for reference extraction. The study employed datasets that included various types of references, such as those from research reports of a particular institution, academic journals, and a combination of academic journal references and non-reference texts. Two deep learning language models, namely RoBERTa+CRF and ChatGPT, were compared to evaluate their performance in automatic extraction. They were used to extract metadata, categorize data types, and separate original text. The research findings showed that the deep learning language models were highly effective, achieving maximum F1-scores of 95.41% for metadata extraction and 98.91% for categorization of data types and separation of the original text. These results provide valuable insights into the use of deep learning language models and different types of datasets for constructing reference databases for research reports including both reference and non-reference texts.

• The Mathematical Education
• /
• v.59 no.3
• /
• pp.255-269
• /
• 2020

In this study, how to instruct the computational estimation of addition and subtraction was considered from the perspective of a 'intended-written-implemented' multi-dimensional curriculum. To this end, the 2015 revised elementary school mathematics curriculum as a intended curriculum and the 2015 revised first~sixth grade textbook as a written curriculum were analyzed with respect to how to instruct the computational estimation of addition and subtraction. As an implemented curriculum, a research study was conducted in relation to the method of instructing teachers about the computational estimation of addition and subtraction. As a result, first, it is necessary to discuss how to develop the ability to estimate and set it as a teaching goal and achievement standard in a separate curriculum to instruct it with learning content. Second, it is necessary to provide an opportunity to learn about various estimation methods by presenting specific activities so that students can learn the estimation itself in a separate operation method. Third, in order to improve the computational estimating ability of addition and subtraction, contents related to the computational estimation need to be included in the achievement criteria, and discussions on the expansion of the areas, and the diversification of the instructing time will be needed.