• Proceedings of the Korea Contents Association Conference
• /
• 2008.05a
• /
• pp.92-95
• /
• 2008

NMF(Non-negative Matrix Factorization) has been proposed as an useful algorithm for feature extraction. Using NMF, we can extract low-dimensional feature vectors. Also, we can find clustering effects in the NMF algorithm. Also, it is reported that the sparse NMF algorithm shows better clustering effects. This paper compares the two approaches in the viewpoint of clustering effects.

• Journal of Korean Institute of Industrial Engineers
• /
• v.25 no.3
• /
• pp.290-297
• /
• 1999

In interior point methods for linear programming, we must solve a linear system with a symmetric positive definite matrix at every iteration, and Cholesky factorization is generally used to solve it. Therefore, if Cholesky factorization is not done successfully, many iterations are needed to find the optimal solution or we can not find it. We studied methods for improving the numerical stability of Cholesky factorization and the accuracy of the solution of the linear system.

• The Journal of the Acoustical Society of Korea
• /
• v.29 no.2
• /
• pp.102-110
• /
• 2010

Music transcription is extracting pitch (the height of a musical note) and rhythm (the length of a musical note) information from audio file and making a music score. In this paper, we decomposed a waveform into frequency and rhythm components using Non-Negative Matrix Factorization (NMF) and Non-Negative Sparse coding (NNSC) which are often used for source separation and data clustering. And using the subharmonic summation method, fundamental frequency is calculated from the decomposed frequency components. Therefore, the accurate pitch of each score can be estimated. The proposed method successfully performed music transcription with its results superior to those of the conventional methods which used either NMF or NNSC.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.52 no.8
• /
• pp.89-96
• /
• 2015

Human visual system has chromatic adaptation to determine the color of an object regardless of illumination, whereas digital camera records illumination and reflectance together, giving the color appearance of the scene varied under different illumination. NMFsc(nonnegative matrix factorization with sparseness constraint) was recently introduced to estimate original object color by using sparseness constraint. In NMFsc, low sparseness constraint is used to estimate illumination and high sparseness constraint is used to estimate reflectance. However, NMFsc has an illumination estimation error for images with large uniform area, which is considered as dominant chromaticity. To overcome the defects of NMFsc, illumination estimation via nonnegative matrix factorization with dominant chromaticity image is proposed. First, image is converted to chromaticity color space and analyzed by chromaticity histogram. Chromaticity histogram segments the original image into similar chromaticity images. A segmented region with the lowest standard deviation is determined as dominant chromaticity region. Next, dominant chromaticity is removed in the original image. Then, illumination estimation using nonnegative matrix factorization is performed on the image without dominant chromaticity. To evaluate the proposed method, experimental results are analyzed by average angular error in the real world dataset and it has shown that the proposed method with 5.5 average angular error achieve better illuminant estimation over the previous method with 5.7 average angular error.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.8
• /
• pp.697-705
• /
• 2009

An unsupervised (blind) method is proposed aiming at extracting rhythmic sources from commercial polyphonic music whose number of channels is limited to one. Commercial music signals are not usually provided with more than two channels while they often contain multiple instruments including singing voice. Therefore, instead of using conventional modeling of mixing environments or statistical characteristics, we should introduce other source-specific characteristics for separating or extracting sources in the under determined environments. In this paper, we concentrate on extracting rhythmic sources from the mixture with the other harmonic sources. An extension of nonnegative matrix factorization (NMF), which is called nonnegative matrix partial co-factorization (NMPCF), is used to analyze multiple relationships between spectral and temporal properties in the given input matrices. Moreover, temporal repeatability of the rhythmic sound sources is implicated as a common rhythmic property among segments of an input mixture signal. The proposed method shows acceptable, but not superior separation quality to referred prior knowledge-based drum source separation systems, but it has better applicability due to its blind manner in separation, for example, when there is no prior information or the target rhythmic source is irregular.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.2
• /
• pp.276-284
• /
• 2003

The preconditioned conjugate gradient method is an efficient iterative solution scheme for large size finite element problems. As preconditioning method, we choose an incomplete Cholesky factorization which has efficiency and easiness in implementation in this paper. The incomplete Cholesky factorization mettled sometimes leads to breakdown of the computational procedure that means pivots in the matrix become minus during factorization. So, it is inevitable that a reduction process fur stabilizing and this process will guarantee robustness of the algorithm at the cost of a little computation. Recently incomplete factorization that enhances robustness through increasing diagonal dominancy instead of reduction process has been developed. This method has better efficiency for the problem that has rotational degree of freedom but is sensitive to parameters and the breakdown can be occurred occasionally. Therefore, this paper presents new method that guarantees robustness for this method. Numerical experiment shows that the present method guarantees robustness without further efficiency loss.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1996.10a
• /
• pp.289-292
• /
• 1996

In Cholesky factorization of the interior point method, dense columns of A matrix make dense Cholesky factor L regardless of sparsity of A matrix. We introduce a method to transform a primal problem to a dual problem in order to preserve the sparsity.

• Journal of the Korean Data and Information Science Society
• /
• v.24 no.2
• /
• pp.333-339
• /
• 2013

This paper discusses a method for getting a basis set of estimable functions of less than full rank linear model. Since model parameters are not estimable estimable functions should be identified for making inferences proper about them. So, it suggests a method of using full rank factorization of model matrix to find estimable functions in easy way. Although they might be obtained in many different ways of using model matrix, the suggested full rank factorization technique could be one of much easier methods. It also discusses how to use projection matrix to identify estimable functions.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.5
• /
• pp.1163-1185
• /
• 2024

With the sharp increase in the volume of literature data, researchers must spend considerable time and energy locating desired papers. A paper recommendation is the means necessary to solve this problem. Unfortunately, the large amount of data combined with sparsity makes personalizing papers challenging. Traditional matrix decomposition models have cold-start issues. Most overlook the importance of information and fail to consider the introduction of noise when using side information, resulting in unsatisfactory recommendations. This study proposes a paper recommendation method (PR-SLSMF) using document-level representation learning with citation-informed transformers (SPECTER) and low-rank and sparse matrix factorization; it uses SPECTER to learn paper content representation. The model calculates the similarity between papers and constructs a weighted heterogeneous information network (HIN), including citation and content similarity information. This method combines the LSMF method with HIN, effectively alleviating data sparsity and cold-start issues and avoiding topic drift. We validated the effectiveness of this method on two real datasets and the necessity of adding side information.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.430-439
• /
• 2007

In this paper, we use non-negative matrix factorization (NMF) to refine the document clustering results. NMF is a dimensional reduction method and effective for document clustering, because a term-document matrix is high-dimensional and sparse. The initial matrix of the NMF algorithm is regarded as a clustering result, therefore we can use NMF as a refinement method. First we perform min-max cut (Mcut), which is a powerful spectral clustering method, and then refine the result via NMF. Finally we should obtain an accurate clustering result. However, NMF often fails to improve the given clustering result. To overcome this problem, we use the Mcut object function to stop the iteration of NMF.