• Smart Structures and Systems
• /
• 제21권6호
• /
• pp.741-749
• /
• 2018

Structural health monitoring (SHM) systems are necessary to achieve smart predictive maintenance and repair planning as well as they lead to a safe operation of mechanical structures. In the context of vibration-based SHM the measured structural responses are employed to draw conclusions about the structural integrity. This usually leads to a mathematically illposed inverse problem which needs regularization. The restriction of the solution set of this inverse problem by using prior information about the damage properties is advisable to obtain meaningful solutions. Compared to the undamaged state typically only a few local stiffness changes occur while the other areas remain unchanged. This change can be described by a sparse damage parameter vector. Such a sparse vector can be identified by employing $L_1$-regularization techniques. This paper presents a novel framework for damage parameter identification by combining sparse solution techniques with an Extended Kalman Filter. In order to ensure sparsity of the damage parameter vector the measurement equation is expanded by an additional nonlinear $L_1$-minimizing observation. This fictive measurement equation accomplishes stability of the Extended Kalman Filter and leads to a sparse estimation. For verification, a proof-of-concept example on a quadratic aluminum plate is presented.

• Journal of Information Processing Systems
• /
• 제9권1호
• /
• pp.157-172
• /
• 2013

There are many recommendation systems available to provide users with personalized services. Among them, the most frequently used in electronic commerce is 'collaborative filtering', which is a technique that provides a process of filtering customer information for the preparation of profiles and making recommendations of products that are expected to be preferred by other users, based on such information profiles. Collaborative filtering systems, however, have in their nature both technical issues such as sparsity, scalability, and transparency, as well as security issues in the collection of the information that becomes the basis for preparation of the profiles. In this paper, we suggest a movie recommendation system, based on the selection of optimal personal propensity variables and the utilization of a secure collaborating filtering system, in order to provide a solution to such sparsity and scalability issues. At the same time, we adopt 'push attack' principles to deal with the security vulnerability of collaborative filtering systems. Furthermore, we assess the system's applicability by using the open database MovieLens, and present a personal propensity framework for improvement in the performance of recommender systems. We successfully come up with a movie recommendation system through the selection of optimal personalization factors and the embodiment of a safe collaborative filtering system.

• 정보과학회 논문지
• /
• 제44권12호
• /
• pp.1333-1337
• /
• 2017

추천 시스템이란 사용자가 좋아할만한 개인화된 상품을 사용자에게 제안하는 것이다. 최근 상품 수의 증가로 추천 시스템의 중요성이 날로 커지고 있지만, 데이터 희소성 문제는 여전히 추천 시스템의 대표적인 문제로 남아있다. 데이터 희소성 문제는 사용자가 전체 상품 중 일부의 상품에만 평점을 부여하여, 사용자와 상품 관계를 정확히 이해하기 힘든 것을 말한다. 이를 해결하기 위해 가장 여러 가지 접근법이 있는 그 중 대표적인 것인 데이터 임퓨테이션이다. 데이터 임퓨테이션은 사용자가 평가하지 않은 상품의 평점을 추론해 평점 행렬에 채우는 방법이다. 하지만 기존 데이터 임퓨테이션 방법은 사용자가 평가하지 않은 상품에 대한 몇 가지 특성을 놓치고 있다. 본 논문에서는 기존 방법의 한계점을 정의하고, 이를 개선하는 방안 3가지를 제안한다.

• Journal of the Korean Data and Information Science Society
• /
• 제25권4호
• /
• pp.807-817
• /
• 2014

본 논문은 최근 많은 관심을 받는 미세먼지 (PM10)의 일별 평균농도에 대해서 전국 16개 시도에서 2008년부터 2011년까지 관측한 다변량 시계열 자료에 대한 연구이다. 다변량 시계열 모형을 이용해서 시간 및 공간에 대한 상관관계를 동시에 고려, 일변량 혹은 특정 지역에 국한해서 분석한 기존의 연구와 차별성을 두었다. 또한 Davis 등 (2013)이 제안한 부분 스펙트럼 일관성 (partial spectral coherence)을 통해 다른 지역간의 상호 의존성을 파악하고 이를 토대로 변수 선택을 통해 희박벡터자기회귀모형 (sVAR; sparse vector autoregressive model)을 적합하는 방법론을 적용하여 고차원 자료 분석의 단점 및 한계를 보완하였으며 예측력 비교를 통해서 sVAR 모형 적합의 타당성을 검증하였다.

• 디지털산업정보학회논문지
• /
• 제11권1호
• /
• pp.171-182
• /
• 2015

The general wavelet transform has profitable property in non-stationary signal analysis specially. The tunable Q-factor wavelet transform is a fully-discrete wavelet transform for which the Q-factor Q and the asymptotic redundancy r, of the transform are easily and independently specified. In particular, the specified parameters Q and r can be real-valued. Therefore, by tuning Q, the oscillatory behavior of the wavelet can be chosen to match the oscillatory behavior of the signal of interest, so as to enhance the sparsity of a sparse signal representation. The TQWT is well suited to fast algorithms for sparsity-based inverse problems because it is a Parseval frame, easily invertible, and can be efficiently implemented. The transform is based on a real valued scaling factor and is implemented using a perfect reconstruction over-sampled filter bank with real-valued sampling factors. The transform is parameterized by its Q-factor and its over-sampling rate, with modest over-sampling rates being sufficient for the analysis/synthesis functions to be well localized. This paper describes filter design of 2D discrete-time wavelet transform for which the Q-factor is easily specified. With the advantage of this transform, perfect reconstruction filter design and implementation for performance improvement are focused in this paper. Hence, the 2D transform can be tuned according to the oscillatory behavior of the image signal to which it is applied. Therefore, application for performance improvement in multimedia communication field was evaluated.

• 디지털산업정보학회논문지
• /
• 제10권3호
• /
• pp.237-247
• /
• 2014

This paper describes a 2D discrete-time wavelet transform for which the Q-factor is easily specified. Hence, the transform can be tuned according to the oscillatory behavior of the image signal to which it is applied. The tunable Q-factor wavelet transform (TQWT) is a fully-discrete wavelet transform for which the Q-factor, Q, of the underlying wavelet and the asymptotic redundancy (over-sampling rate), r, of the transform are easily and independently specified. In particular, the specified parameters Q and r can be real-valued. Therefore, by tuning Q, the oscillatory behavior of the wavelet can be chosen to match the oscillatory behavior of the signal of interest, so as to enhance the sparsity of a sparse signal representation. The TQWT is well suited to fast algorithms for sparsity-based inverse problems because it is a Parseval frame, easily invertible, and can be efficiently implemented. The TQWT can also be used as an easily-invertible discrete approximation of the continuous wavelet transform. The transform is based on a real valued scaling factor (dilation-factor) and is implemented using a perfect reconstruction over-sampled filter bank with real-valued sampling factors. The transform is parameterized by its Q-factor and its oversampling rate (redundancy), with modest oversampling rates (e. g. 3-4 times overcomplete) being sufficient for the analysis/synthesis functions to be well localized. Therefore, This method services good performance in image processing fields.

• 응용통계연구
• /
• 제29권1호
• /
• pp.27-39
• /
• 2016

본 논문은 다차원의 시계열 자료 분석에서 효율적인 희박벡터자기회귀모형에서의 모수 추정에 대해서 연구한다. 희박벡터자기회귀모형은 영에 가까운 계수를 정확이 영으로 둠으로써 희박성을 확보한다. 따라서 변수 선택과 모수 추정을 한꺼번에 할 수 있는 lasso를 이용한 방법론을 희박벡터자기회귀모형의 추정에 쓸 수 있다. 하지만 Davis 등(2015)에서는 모의실험을 통해 일반적인 lasso의 경우 영이아닌 계수를 참값보다 훨씬 더 많이 찾아 희박성에 약점이 있음을 보고하였다. 이에 따라 본 연구는 희박벡터자기회귀모형에 adaptive lasso를 이용하면 일반 lasso보다 희박성을 비롯한 전반적인 모수의 추정이 매우 유의하게 개선됨을 보인다. 또한 adaptive lasso에서 쓰이는 튜닝 모수들에 대한 선택도 아울러 논의한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권12호
• /
• pp.6043-6062
• /
• 2019

Low-rank matrix decomposition has shown its capability in many applications such as image in-painting, de-noising, background reconstruction and defect detection etc. In this paper, we consider the texture background of rail track images and the sparse foreground of the defects to construct a low-rank matrix decomposition model with block sparsity for defect inspection of rail tracks, which jointly minimizes the nuclear norm and the 2-1 norm. Similar to ADM, an alternative method is proposed in this study to solve the optimization problem. After image decomposition, the defect areas in the resulting low-rank image will form dark stripes that horizontally cross the entire image, indicating the preciselocations of the defects. Finally, a two-stage defect extraction method is proposed to locate the defect areas. The experimental results of the two datasets show that our algorithm achieved better performance compared with other methods.

• 한국전자파학회논문지
• /
• 제27권4호
• /
• pp.388-394
• /
• 2016

안테나 배열을 통한 방향 탐지는 여러 분야에서 활발하게 이루어지고 있는 연구 분야이다. Beamforming, Capon's method, maximum likelihood(ML), MUSIC 등과 같은 방향 탐지 알고리즘이 대표적이다. 최근 방향 탐지 이론은 압축센싱기법을 이용하여 신호의 희소도를 이용한 방법의 연구가 수행되고 있다. 본 논문에서는 그 중 하나인 신호의 데이터 행렬을 fitting하는 L1-SVD 알고리즘의 성능을 알아보기 위해 MUSIC 알고리즘과 비교하여 장단점을 알아본다.

• Journal of Information Processing Systems
• /
• 제16권6호
• /
• pp.1343-1358
• /
• 2020

In the stagewise arithmetic orthogonal matching pursuit algorithm, the weak threshold used in sparsity estimation is determined via maximum iterations. Different maximum iterations correspond to different thresholds and affect the performance of the algorithm. To solve this problem, we propose an improved variable weak threshold based on the stagewise arithmetic orthogonal matching pursuit algorithm. Our proposed algorithm uses the residual error value to control the weak threshold. When the residual value decreases, the threshold value continuously increases, so that the atoms contained in the atomic set are closer to the real sparsity value, making it possible to improve the reconstruction accuracy. In addition, we improved the generalized Jaccard coefficient in order to replace the inner product method that is used in the stagewise arithmetic orthogonal matching pursuit algorithm. Our proposed algorithm uses the covariance to replace the joint expectation for two variables based on the generalized Jaccard coefficient. The improved generalized Jaccard coefficient can be used to generate a more accurate calculation of the correlation between the measurement matrixes. In addition, the residual is more accurate, which can reduce the possibility of selecting the wrong atoms. We demonstrate using simulations that the proposed algorithm produces a better reconstruction result in the reconstruction of a one-dimensional signal and two-dimensional image signal.