• ETRI Journal
• /
• 제37권2호
• /
• pp.317-327
• /
• 2015

A reliable analysis of consumer preference from a large amount of purchase data acquired in real time and an accurate customer characterization technique are essential for successful direct marketing campaigns. In this study, an optimal segmentation of post office customers in Korea is performed using a subspace projection-based clustering method to generate an accurate customer characterization from a high-dimensional census dataset. Moreover, a traditional temporal mining method is extended to an algorithm using the MapReduce framework for a consumer preference analysis. The experimental results show that it is possible to use parallel mining through a MapReduce-based algorithm and that the execution time of the algorithm is faster than that of a traditional method.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2013년도 춘계학술발표대회
• /
• pp.908-910
• /
• 2013

Customers of different electricity consumer types have different daily load shapes in the manner of different characteristics. Therefore, maximally capture load shape variability are desirable in load flow analysis. And most of time, such load shape variability can be found in the particular subspace of load diagrams. Therefore, in this paper, we are using subspace projection method to capture the emerging subspaces of load diagrams which maximize the difference between particular load shapes in different group of customers. As the result, subspace projection method can be used in load profiling and the performance is good as traditional approaches.

• 전자공학회논문지
• /
• 제51권9호
• /
• pp.148-156
• /
• 2014

MVDR(minimum variance distortionless response) 적응 어레이에서 조향벡터(steering vector)에 에러가 있으면 원하는 신호(desired signal)도 감쇠되어 성능이 심하게 저하될 수 있다. 본 논문에서는 이러한 에러에 대응할 수 있는 계산이 간편한 기법을 제안한다. 제안한 방법에서는 DCB(doubly constrained beamforming) 원리에 기초한 최소화 문제의 해 벡터를 구하고, 이벡터를 부공간에 투사하여 새로운 조향벡터로 사용한다. 최소화 문제의 해결과 부공간 투사에 필요한 주 고유쌍(principal eigenpairs)은 PASTd(projection approximation subspace tracking with deflation)를 변형한 MPASTd(modified PASTd)에 의거하여 직접 상관행렬(correlation matrix)을 추정함이 없이 수신 데이터로부터 구해진다. 그리고 고유쌍 계산에 있어, 기존에 알려진 MPASTd를 개선해서 계산량을 절감하면서 효과적으로 구하는 방법을 제시한다. 제안한 적응어레이 기법은 상관행렬을 추정하고 이를 고유분해(eigendecomposition)하는 기존방식보다 계산량을 크게 줄이고 우수한 성능을 가질 수 있다.

• 대한전자공학회논문지SP
• /
• 제47권6호
• /
• pp.162-170
• /
• 2010

원하는 신호의 도래방향에 관한 정보를 이용하여 적응 어레이는 이 방향으로 빔 이득을 유지하면서 간섭신호를 제거한다. 신호 부공간에서 가중벡터를 조정하면 전체 공간에서 조정하는 방식에 비해 빠른 수렴속도를 가지며, 도래각 정보에서의 에러에 강인한 특성을 가진다. 그러나 공분산 행렬의 고유분해가 필요하고 이에 따른 계산이 복잡하다. 본 논문에서는 PASTd(projection approximation subspace tracking with deflation) 방식에 기초하여 계산이 간단한 신호 부공간에 기초한 적응어레이를 제시한다. 제시된 방식은 고유벡터가 직교하도록 원래의 PASTd를 변형해서 사용하고 있고, 직접 고유분해하는 방식과 동일한 성능을 가지면서 계산량을 크게 감소시킬 수 있다. 또한 신호 부공간 어레이의 SINR(signal-to-interference plus noise ratio)성능을 이론적으로 분석하여, 이의 동작특성을 고찰하였다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.737-741
• /
• 2009

This paper addresses the factorization method to estimate the projective structure of a scene from feature (points) correspondences over images with occlusions. We propose both a column and a row space approaches to estimate the depth parameter using the subspace constraints. The projective depth parameters are estimated by maximizing projection onto the subspace based either on the Joint Projection matrix (JPM) or on the the Joint Structure matrix (JSM). We perform the maximization over significant observation and employ Tardif's Camera Basis Constraints (CBC) method for the matrix factorization, thus the missing data problem can be overcome. The depth estimation and the matrix factorization alternate until convergence is reached. Result of Experiments on both real and synthetic image sequences has confirmed the effectiveness of our proposed method.

• Communications for Statistical Applications and Methods
• /
• 제23권2호
• /
• pp.93-103
• /
• 2016

In the paper, we discuss dimension reduction of predictors ${\mathbf{X}}{\in}{{\mathbb{R}}^p}$ in a regression of $Y{\mid}{\mathbf{X}}$ with a notion of sufficiency that is called sufficient dimension reduction. In sufficient dimension reduction, the original predictors ${\mathbf{X}}$ are replaced by its lower-dimensional linear projection without loss of information on selected aspects of the conditional distribution. Depending on the aspects, the central subspace, the central mean subspace and the central $k^{th}$-moment subspace are defined and investigated as primary interests. Then the relationships among the three subspaces and the changes in the three subspaces for non-singular transformation of ${\mathbf{X}}$ are studied. We discuss the two conditions to guarantee the existence of the three subspaces that constrain the marginal distribution of ${\mathbf{X}}$ and the conditional distribution of $Y{\mid}{\mathbf{X}}$. A general approach to estimate them is also introduced along with an explanation for conditions commonly assumed in most sufficient dimension reduction methodologies.

• Journal of International Society for Simulation Surgery
• /
• 제2권2호
• /
• pp.64-66
• /
• 2015

Purpose : In this paper we proposed cascade feature filter and projection method for rapid human face recognition for the large-scale high-dimensional face database. Materials and Methods : The relevant features are selected from the large feature set using Fast Correlation-Based Filter method. After feature selection, project them into discriminant using Principal Component Analysis or Linear Discriminant Analysis. Their cascade method reduces the time-complexity without significant degradation of the performance. Results : In our experiments, the ORL database and the extended Yale face database b were used for evaluation. On the ORL database, the processing time was approximately 30-times faster than typical approach with recognition rate 94.22% and on the extended Yale face database b, the processing time was approximately 300-times faster than typical approach with recognition rate 98.74 %. Conclusion : The recognition rate and time-complexity of the proposed method is suitable for real-time face recognition system on the large-scale high-dimensional face database.

• Journal of applied mathematics & informatics
• /
• 제40권3_4호
• /
• pp.669-674
• /
• 2022

In this paper, we shall consider some properties of the metric projection as a set valued mapping. For a set G in a metric space E, the mapping 𝜋_G; x → 𝜋_G(x) of E into 2^G is called set valued metric projection of E onto G. We investigated the properties related to the projection P_S(·)(·) and 𝜋_S(·)(·) as one-sided best simultaneous approximations.

• The Journal of the Acoustical Society of Korea
• /
• 제20권1E호
• /
• pp.7-13
• /
• 2001

In this paper we propose a new subspace tracking algorithm based on the PASTd (Projection Approximation Subspace Tracking with deflation). The algorithm is obtained via introducing the variable forgetting factor which adapts itself to the time-varying subspace environments. The tracking capability of the proposed algorithm is demonstrated by computer simulations in an abruptly changing DOA scenario. The estimation results of the variable forgetting factor PASTd(VFF-PASTd) outperform those of the PASTd in the nonstationary case as well as in the stationary case.

• Korean Chemical Engineering Research
• /
• 제48권4호
• /
• pp.475-482
• /
• 2010

화상분석(image analysis)을 이용하여 제품의 외관(外觀) 품질을 정량적으로 추정할 수 있는 소프트 센서를 설계하고, 이를 제품의 품질 모니터링에 적용하는 연구를 수행하였다. 여기에 사용된 방법론은 크게 다음의 세 단계로 구성되어 있다: (1) 웨이블릿 변환(wavelet transform)을 이용한 화상으로부터의 질감(texture) 특징 추출, (2) 추출된 질감특징의 부공간 투영(projection on subspace)을 통한 제품 외관의 추정, 그리고 (3) 질감특징의 잠재변수(latent variables) 즉, 외관의 수치적 추정치를 목적에 맞게 사용. 이 방법에서는 제품의 외관을 서로 다른 불연속적인 부류로의 분류 보다는, 연속적인 외관 변화를 일관적이고 정량적으로 추정하는데 초점을 두고자 한다. 이 방법은 인조대리석 외관의 수치적 추정과 품질 모니터링 적용사례를 통해 설명되었다.