• Proceedings of the Korea Information Processing Society Conference
• /
• 2004.11a
• /
• pp.583-586
• /
• 2004

본 논문에서는 자율학습의 속성을 가지는 주요성분분석을 이용한 얼굴인식 기법을 제안하였다. 이는 대용량의 입력 데이터를 통계적으로 독립인 특징들의 집합으로 변환시켜 중복신호를 제거하는 특성을 가지는 주요성분분석의 우수한 속성을 이용한 것이다. 제안된 기법을 Yale 얼굴영상 데이터베이스로부터 선택된 20개의 $320{\ast}243$ 픽셀의 영상을 대상으로 시뮬레이션한 결과, 주요성분의 개수에 따른 압축성능과 city-block, Euclidian, 그리고 negative angle(cosine)의 거리척도에 따른 인식에서의 분류성능에서 우수한 성능이 있음을 확인할 수 있었다.

• Journal of KIISE:Databases
• /
• v.35 no.2
• /
• pp.182-191
• /
• 2008

Most of previous works for skyline queries have focused only on static attributes of target objects. With the advance in mobile applications, however, the need of continuous skyline queries for moving objects has been increasing. Even though several techniques to process continuous skyline queries have been proposed recently, they cannot process subspace queries, which use only the subset of attribute dimensions. Therefore it is not feasible to utilize those methods for mobile applications which must consider moving objects and subspaces simultaneously. In this paper, we propose a dominant object-based pruning method to compute subspace skyline of moving objects efficiently at query time and present the experimental results to show the effectiveness of the proposed method.

• Journal of Intelligence and Information Systems
• /
• v.4 no.1
• /
• pp.11-24
• /
• 1998

이 논문에서는 상황에 적합한 정보들을 선택하고 이를 해당되는 GUI (Graphic User Interface) 객체를 통해 사용자에게 제시하기 위한 규칙 기반의 지능적 사용적 인터페이스 구조를 제안하고, 이를 이용하여 실제로 군사 분야 정보 융합 시스템의 그래픽 사용자 인터페이스를 개발하였다. 군사분야와 같이 복잡하고 급변하는 상황에서는 상황에 따라 필요한 정보가 다를 뿐만 아니라, 불필요한 정보의 제시로 인해 오히려 사용자의 판단에 혼란을 초래할 수 있다. 이 논문에서는 이러한 점을 고려하여, 제시할 정보와 GUI 객체의 적절한 대응뿐만 아니라, 상황에 따라 사용자에게 필요한 정보를 해당분야의 전문가로부터 획득한 규칙에 의해 선택적으로 제시하도록 하기 위한 GUI 구조를 제안하였다. 개발한 GUI의 구조는 입력 해석, 제시할 정보의 선택, GUI 객체의 선택, GUI객체의 생성 및 속성 기정, GUI 객체의 배치 및 출력 단계들로 구성되며, 각각의 단계는 규칙에 의해 처리된다. 제안된 구조에 의해 공중 작전 수립을 위한 정보 융합 시스템의 사용자 인터페이스를 개발한 결과, 다양한 매체를 통해 많은 데이터가 들어오는 상황에서 필요한 정보만을 선택하여 적절히 제시해 줌으로써 사용자의 신속한 상황 판단과 결정 수립을 지원할 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.05a
• /
• pp.400-401
• /
• 2014

본 논문에서는 이동 에드혹 네트워크(Mobile Ad hoc Network: MANET)에서의 상황인식 기반의 스케쥴링 기법인 DDV(Dynamic Direction Vector)-hop알고리즘을 제안한다. 기존 MANET에서는 노드의 이동성으로 인한 동적 네트워크 토폴리지, 네트워크 확장성 결여의 대한 취약성을 지니고 있다. 본 논문에서는 계층적 클러스터 단위의 동적인 토폴로지에서 노드가 이동하는 방향성 및 속도에 대한 노드의 이동 속성 정보를 고려하여 클러스터를 생성 및 유지하는 DDV-hop 알고리즘을 제안한다. 제안된 알고리즘은 클러스터 헤드노드를 기준으로 클러스터 멤버노드의 방향성 및 속도의 속성 정보를 비교하여 유사한 노드간 클러스터링을 구성하고, 이로부터 헤드노드를 선택하는 방법이다. 실험결과, 제안하는 알고리즘이 네트워크의 부하를 감소시키고 네트워크 토폴로지를 안정적으로 유지할 수 있음을 확인하였다.

• Journal of the Korea Convergence Society
• /
• v.8 no.3
• /
• pp.215-223
• /
• 2017

The purpose of the present study is to identify the differences of customer preference in online and offline used car business. Conjoint analysis is used for examining the attributes of used car choices. As a result, the order of importance in real used car market is as follows: brand, design, price, model, mileage. Whereas the order of importance in online used car market is as follows: brand, trust, price, web design, accident. Accordingly, the author suggested that customer preferences depend on the path people are approaching and the attributes of preference vary in online and real stores. For example, trust and accident are important attributes in online market in comparison with real market. Used car market is increasing and becoming important. The authorities and practitioners need to understand used car market and establish the related policies.

• Information Systems Review
• /
• v.22 no.4
• /
• pp.21-39
• /
• 2020

Recently, deep learning has shown high performance in various applications such as pattern analysis and image classification. Especially known as a difficult task in the field of machine learning research, stock market forecasting is an area where the effectiveness of deep learning techniques is being verified by many researchers. This study proposed a deep learning Convolutional Neural Network (CNN) model to predict the direction of stock prices. We then used the feature selection method to improve the performance of the model. We compared the performance of machine learning classifiers against CNN. The classifiers used in this study are as follows: Logistic Regression, Decision Tree, Neural Network, Support Vector Machine, Adaboost, Bagging, and Random Forest. The results of this study confirmed that the CNN showed higher performancecompared with other classifiers in the case of feature selection. The results show that the CNN model effectively predicted the stock price direction by analyzing the embedded values of the financial data

• The KIPS Transactions:PartB
• /
• v.13B no.7 s.110
• /
• pp.679-688
• /
• 2006

Protein-protein interaction data obtained from high-throughput experiments includes high false positives. In this paper, we introduce a new protein-protein interaction reliability verification system. The proposed system integrates various biological features related with protein-protein interactions, and then selects the most relevant and informative features among them using a feature selection method. To assess the reliability of each protein-protein interaction data, the system construct a classifier that can distinguish true interacting protein pairs from noisy protein-protein interaction data based on the selected biological evidences using a classification technique. Since the performance of feature selection methods and classification techniques depends heavily upon characteristics of data, we performed rigorous comparative analysis of various feature selection methods and classification techniques to obtain optimal performance of our system. Experimental results show that the combination of feature selection method and classification algorithms provide very powerful tools in distinguishing true interacting protein pairs from noisy protein-protein interaction dataset. Also, we investigated the effects on performances of feature selection methods and classification techniques in the proposed protein interaction verification system.

• Journal of the Computational Structural Engineering Institute of Korea
• /
• v.35 no.5
• /
• pp.267-275
• /
• 2022

Using a nonconforming mesh in enrichment methods results in several numerical issues induced by discontinuities and singularities found within the solution spaces, including the computational overhead during integration. In this study, we present a novel enrichment technique based on the selective expansion technique of moment fitting (Düster and Allix, 2020). In particular, two modifications are proposed to address the inefficiency during the integration process. First, a feedforward artificial neural network is introduced to correlate the implicit functions and integration moments. Through numerical examples, it is shown that the efficiency of the method is greatly improved when compared with existing expansion techniques, whereas the solution accuracy is maintained. Additionally, the finite element and domain representation grids are separated, which in turn improves the solution accuracy even for coarse mesh conditions.

• The Journal of the Korea Contents Association
• /
• v.15 no.5
• /
• pp.66-77
• /
• 2015

The main objective of this study was to identify the difference in the perceptions of the product attributes and the expected benefits for consumers and professionals in the performing arts industry. First, the Delphi method by using email and telephone was used to explore the professionals' perceptions of the attributes of performing arts and the expected benefits. Second, on-line consumer survey was used to explore the consumer perceptions of the attributes of performing arts and the expected benefits. Finally, this study tried to draw some difference of perceptions from both professionals and consumers. Contents of the arts, location of a theater, ticket price, relief of stress, and improvement of lifestyle were found to be commonly important factors for both parties. Among them, statistical differences between two groups were confirmed in the factors such as the reputations of the cast members and theatres, locations of theatres, and time duration of a performance.

• The Journal of the Korea Contents Association
• /
• v.12 no.4
• /
• pp.438-447
• /
• 2012

The objectives of this research are: to provide fundamental data for tourism policy and investigate medical tourism attributes by IPA(Importance-Performance Analysis) to attract Japanese medical tourists to Korea. Utilizing Japanese medical tourists as subjects of inquiry, this study evaluated the importance of medical tourism selection attributes and the rate at which these attributes were performed. Consequently, this study deduced 'convenience, service quality, differentiation, approximation' factors by factor analysis as well as the ranking among each selection attributes using T-test. The importance-performance analysis showed 'Concentrate Here' in quadrant I, 'Keep up the Good Work' in quadrant II, 'Low Priority' in quadrant III and 'Possible Overkill' in quadrant IV. The results show that "Concentrate Here" part in quadrant I is reservation procedure rapidity and information system simplicity. This means that these attributes are very important but they indicate low performance. Thus, concentration among tourism managers should be focused along these attributes. The study, then, suggests that employee who speaks Japanese fluently should be arranged during reservation procedure. Furthermore, Japanese class should become mandatory during basic employee training to improve reservation procedure rapidity. Moreover, clinic websites should concentrate on accessibility of Japanese users in such a way that they can consult their needs on line. Finally, managers should find diverse ways to bring about good service rapidly and conveniently.