• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.2
• /
• pp.832-854
• /
• 2019

Action recognition has been studied in computer vision field for years. We present an effective approach to recognize actions using a dimension reduction method, which is applied as a crucial step to reduce the dimensionality of feature descriptors after extracting features. We propose to use sparse matrix and randomized kd-tree to modify it and then propose modified Local Fisher Discriminant Analysis (mLFDA) method which greatly reduces the required memory and accelerate the standard Local Fisher Discriminant Analysis. For feature encoding, we propose a useful encoding method called mix encoding which combines Fisher vector encoding and locality-constrained linear coding to get the final video representations. In order to add more meaningful features to the process of action recognition, the convolutional neural network is utilized and combined with mix encoding to produce the deep network feature. Experimental results show that our algorithm is a competitive method on KTH dataset, HMDB51 dataset and UCF101 dataset when combining all these methods.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.4
• /
• pp.535-542
• /
• 1995

In this paper we present two preconditioners for solving large sparse linear systems arising from elliptic partial differential equations on massively parallel machines, such as the CM-5. Most massively parallel machines do heavily rely on the message-passing for the interprocessor communications. but according to the current manufacturing standards the cost of communications is very high compared to that of floating point arithmetic computations. Due to this we need an algorithm which minimizes the amount of interprocessor communication on the massively parallel machines. We will show that Block SOR(Successive Over Relaxation) method coupled with the multi-coloring technique is one of such preconditioner on the massively parallel machines, by conducting experiments in the CM-5. Also, we implemented the ADI(Alternation Direction Implicit) method in the CM-5, which has been conventionally one of the most powerful parallel preconditioner. Our experiment shows that Block SOR method coupled with the multi-coloring technique could yield a speedup with 50% efficiency with the range of number of processors form 16 to 512 for a matrix with dimension 512x512. On the other hand, the ADI method shows a very poor performance.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.8
• /
• pp.3498-3511
• /
• 2016

We investigate the channel state information (CSI) in multi-input multi-output (MIMO) cooperative networks that employ the amplify-and-forward transmission scheme. Least squares and expectation conditional maximization have been proposed in the system. However, neither of these two approaches takes advantage of channel sparsity, and they cause estimation performance loss. Unlike linear channel estimation methods, several compressed channel estimation methods are proposed in this study to exploit the sparsity of the MIMO cooperative channels based on the theory of compressed sensing. First, the channel estimation problem is formulated as a compressed sensing problem by using sparse decomposition theory. Second, the lower bound is derived for the estimation, and the MIMO relay channel is reconstructed via compressive sampling matching pursuit algorithms. Finally, based on this model, we propose a novel algorithm so called sparsity adaptive expectation maximization (SAEM) by using Kalman filter and expectation maximization algorithm so that it can exploit channel sparsity alternatively and also track the true support set of time-varying channel. Kalman filter is used to provide soft information of transmitted signals to the EM-based algorithm. Various numerical simulation results indicate that the proposed sparse channel estimation technique outperforms the previous estimation schemes.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.2
• /
• pp.205-211
• /
• 2012

The retrieved documents have to be transformed into proper data structure for the clustering algorithms of statistics and machine learning. A popular data structure for document clustering is document-term matrix. This matrix has the occurred frequency value of a term in each document. There is a sparsity problem in this matrix because most frequencies of the matrix are 0 values. This problem affects the clustering performance. The sparseness of document-term matrix decreases the performance of clustering result. So, this research uses the factor score by factor analysis to solve the sparsity problem in document clustering. The document-term matrix is transformed to document-factor score matrix using factor scores in this paper. Also, the document-factor score matrix is used as input data for document clustering. To compare the clustering performances between document-term matrix and document-factor score matrix, this research applies two typed matrices to self organizing map (SOM) clustering.

• International Journal of Quality Innovation
• /
• v.7 no.2
• /
• pp.69-82
• /
• 2006

Organisations worldwide strive to develop their management systems for business functions, ranging from quality and environment to safety, information security and social responsibility. During the latest decade a considerable amount of these efforts has been concentrated on introducing and applying standards such as the ISO 9001 and ISO 14001. The need for Integrated Management Systems (IMS) often arises as a result of decisions to implement Environmental Management System (EMS) and/or an occupational health and safety management system in addition to a Quality Management System (QMS). At the end of 2003, approximately 3200 organisations in Sweden had an ISO 9001 certificate, and approximately 3400 organisations had a certificate based on an EMS. Dealing with separate management systems and ensuring that they align with the organisation's strategies and goals, has proved difficult. Owing to the large number of organisations certified according to multiple types of systems, an increasing number of organisations are establishing IMS. There are examples of companies, which chose to integrate EMS and QMS into a co-ordinated implementation approach, and although sparse, the research within this area indicates potential benefits of using an integrated approach. This paper presents both a theoretical and an empirical investigation with the aim to elucidate problems related to the integration of management systems. Furthermore, the paper will present recommendations for succeeding in such integrations and, hence, contributing to an increased understanding on how IMSs should be designed and implemented.

• Journal of Internet Computing and Services
• /
• v.25 no.1
• /
• pp.49-56
• /
• 2024

There have been emerging many use-cases applying recommendation systems especially in online platform. Although the performance of recommendation systems is affected by a variety of factors, selecting appropriate features is difficult since most of recommendation systems have sparse data. Conventional matrix factorization (MF) method is a basic way to handle with problems in the recommendation systems. However, the MF based scheme cannot reflect non-linearity characteristics well. As deep learning technology has been attracted widely, a deep neural network (DNN) framework based collaborative filtering (CF) was introduced to complement the non-linearity issue. However, there is still a problem related to feature embedding for use as input to the DNN. In this paper, we propose an effective method using singular value decomposition (SVD) based feature embedding for improving the DNN performance of recommendation algorithms. We evaluate the performance of recommendation systems using MovieLens dataset and show the proposed scheme outperforms the existing methods. Moreover, we analyze the performance according to the number of latent features in the proposed algorithm. We expect that the proposed scheme can be applied to the generalized recommendation systems.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.86-89
• /
• 2003

To model a numerical problem space under the limitation of available data, we need to extract sparse but key points from the space and to efficiently approximate the space with them. This study proposes a sampling method based on the search process of genetic algorithm and a space modeling method based on least-squares approximation using the summation of Gaussian functions. We conducted simulations to evaluate them for several kinds of problem spaces: DeJong's, Schaffer's, and our original one. We then compared the performance between our sampling method and sampling at regular intervals and that between our modeling method and modeling using a polynomial. The results showed that the error between a problem space and its model was the smallest for the combination of our sampling and modeling methods for many problem spaces when the number of samples was considerably small.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2000.11a
• /
• pp.480-483
• /
• 2000

공학의 많은 응용분야에서 큰 회소 행렬(Large Sparse Matrices)에 대한 가장 작거나 또는 가장 큰 고유치(Eigenvalues)들을 요구하게 되는데, 이때 많이 이용되는 것은 Krylov Subspace로의 Projection방법이다. 대칭 행렬에 대해서는 Lanczos방법을, 비대칭 행렬에 대해서는 Biorhtogonal Lanczos방법을 이용할 수 있다. 이러한 기존의 알고리즘들은 새롭게 제안되는 병렬처리 시스템에서 효과적이지 못하다. 많은 프로세서를 가지는 병렬처리 컴퓨터 중에서도 분산 기억장치 시스템(Distributed Memory System)에서는 프로세서들 사이의 Data Communication에 필요한 시간을 줄이도록 해야한다. 본 논문에서는 기존의 Lanczos 알고리즘을 수정함으로써, 알고리즘의 동기점(Synchronization Point)을 줄이고 병렬화를 위한 입상(Granularity)을 증가시켜서 MPP인 Cray T3E에서 Data Communication에 필요한 시간을 줄인다. 많은 프로세서를 사용하는 경우 수정된 알고리즘이 기존의 알고리즘에 비해 더 나은 speedup을 보여준다.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.1
• /
• pp.52-60
• /
• 2008

This research proposes a new strategy where documents are encoded into string vectors for text categorization and modified versions of SVM to be adaptable to string vectors. Traditionally, when the traditional version of SVM is used for pattern classification, raw data should be encoded into numerical vectors. This encoding may be difficult, depending on a given application area of pattern classification. For example, in text categorization, encoding full texts given as raw data into numerical vectors leads to two main problems: huge dimensionality and sparse distribution. In this research, we encode full texts into string vectors, and apply the modified version of SVM adaptable to string vectors for text categorization.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.499-501
• /
• 2004

웹 로그, 바이오정보학 둥 여러 분야에서 다양한 형태의 결측치가 발생하여 학습 데이터를 희소하게 만든다. 결측치는 주로 전처리 과정에서 조건부 평균이나 나무 모형과 같은 기본적인 Imputation 방법을 이용하여 추정된 값에 의해 대체되기도 하고 일부는 제거되기도 한다. 특히, 결측치 비율이 매우 크게 되면 기존의 결측치 대체 방법의 정확도는 떨어진다. 또한 데이터의 결측치 비율이 증가할수록 사용 가능한 Imputation 방법들의 수는 극히 제한된다. 이러한 문제점을 해결하기 위하여 본 논문에서는 Vapnik의 Support Vector Regression을 데이터 전처리 과정에 알맞게 변형한 Support Vector Regression을 제안하여 이러한 문제점들을 해결하였다. 제안 방법을 통하여 결측치의 비율이 상당히 큰 희소 데이터의 전처리도 가능하게 되었다. UCI machine learning repository로부터 얻어진 데이터를 이용하여 제안 방법의 성능을 확인하였다.