• The Journal of Engineering Geology
• /
• v.28 no.3
• /
• pp.465-473
• /
• 2018

The application of singular value decomposition (SVD) filtering is examined for attenuation of the ground-roll in land seismic data. Prior to the SVD computation to seek singular values containing the highly correlatable reflection energy, processing steps such as automatic gain control, elevation and refraction statics, NMO correction, and residual statics are performed to enhance the horizontal correlationships and continuities of reflections. Optimal parameters of SVD filtering are effectively chosen with diagnostic display of inverse NMO (INMO) corrected CSP (common shot point) gather. On the field data with dispersion of ground-roll overwhelmed, continuities of reflection events are much improved by SVD filtering than f-k filtering by eliminating the ground-roll with preserving the low-frequency reflections. This is well explained in the average amplitude spectra of the f-k and SVD filtered data. The reflectors including horizontal layer of the reservoir are much clearer on the stack section, with laminated events by SVD filtering and subsequent processing steps of spiking deconvolution and time-variant spectral whitening.

• Geophysics and Geophysical Exploration
• /
• v.22 no.4
• /
• pp.177-185
• /
• 2019

Ground roll is a surface wave which is usually observed in the land seismic data. It is one of the typical coherent noise. During the reflection data processing, ground roll is removed because it is considered as noise. This removal process often causes the loss of reflection signals if the ground roll overlaps reflection signals. In this study, we look over Karhunen-Loeve Transform (KLT) and analyze its effects to suppress the ground roll appropriately while reducing the reflection loss. Numerical tests in homogeneous elastic media show that the ground roll has been properly rejected. However, the field data application reveals that there is no significant suppression of ground roll when compared to band-pass filtering. This can be considered that it is hard to calculate horizontally aligned gathers in the field data because the ground roll contains a wide range of frequency bands. On the contrary, the result of singular value decomposition (SVD) filtering shows that the ground roll has been significantly reduced. It is thought that the SVD filtering performs better in the ground roll suppression than KLT because it is easy to calculate the horizontally aligned gathers in the SVD filtering.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2018.10a
• /
• pp.493-496
• /
• 2018

추천 시스템은 다양한 분야에 적용되는 기술로서 활발한 연구가 진행되고 있고 기존 추천 시스템의 성능을 높이기 위해서 더욱 개인화된 차세대 추천 시스템의 필요성이 대두되고 있다. 본 논문은 하이퍼 개인화 범주에 속하는 사후 필터링기법을 사용한 실시간 상황 인식 추천 시스템을 제안한다. 실시간 상황 인식 추천 시스템은 사용자 행동과 계속적인 동기화로 현재 상황에 가장 적합한 추천 목록을 생성하기 때문에 사용자 기반 협업 필터링 (User Based Collaborative Filtering), 콘텐츠 기반 필터링(Content-based Filtering), 특이값 분해(Singular Value Decomposition)보다 훨씬 미래 지향적인 추천 시스템이다.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.6
• /
• pp.19-28
• /
• 2021

Nowadays, interest in health care is increasing due to Coronavirus (COVID-19), and a lot of people are doing home training as there are more difficulties in using fitness centers and public facilities that are used together. In this paper, we propose a personalized exercise recommendation algorithm using personalized propensity information to provide more accurate and meaningful exercise recommendation to home training users. Thus, we classify the data according to the criteria for obesity with a k-nearest neighbor algorithm using personal information that can represent individuals, such as eating habits information and physical conditions. Furthermore, we differentiate the exercise dataset by the level of exercise activities. Based on the neighborhood information of each dataset, we provide personalized exercise recommendations to users through a dimensionality reduction algorithm (SVD) among model-based collaborative filtering methods. Therefore, we can solve the problem of data sparsity and scalability of memory-based collaborative filtering recommendation techniques and we verify the accuracy and performance of the proposed algorithms.

• The Journal of Bigdata
• /
• v.2 no.1
• /
• pp.41-48
• /
• 2017

Recently, the popularity of the recommendation system and the evaluation of the performance of the algorithm of the recommendation system have become important. In this study, we used modeling and RMSE to verify the effectiveness of various algorithms in movie data. The data of this study is based on user-based collaborative filtering using Pearson correlation coefficient, item-based collaborative filtering using cosine correlation coefficient, and item-based collaborative filtering model using singular value decomposition. As a result of evaluating the scores with three recommendation models, we found that item-based collaborative filtering accuracy is much higher than user-based collaborative filtering, and it is found that matrix recommendation is better when using matrix decomposition.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.8
• /
• pp.39-48
• /
• 2023

This study was conducted to provide a solution to the problem of sizing errors occurring in online purchases due to discrepancies and non-standardization in clothing sizes. This paper discusses an implementation approach for a machine learning-based recommender system capable of providing personalized sizes to online consumers. We trained multiple validated collaborative filtering algorithms including Non-Negative Matrix Factorization (NMF), Singular Value Decomposition (SVD), k-Nearest Neighbors (KNN), and Co-Clustering using purchasing data derived from online commerce and compared their performance. As a result of the study, we were able to confirm that the NMF algorithm showed superior performance compared to other algorithms. Despite the characteristic of purchase data that includes multiple buyers using the same account, the proposed model demonstrated sufficient accuracy. The findings of this study are expected to contribute to reducing the return rate due to sizing errors and improving the customer experience on e-commerce platforms.

• Proceedings of the Korea Society for Industrial Systems Conference
• /
• 2002.11a
• /
• pp.407-411
• /
• 2002

Ternary sequences are digital codes consisting of discrete values -1, 0, and 1 only. They are advantageous in that the correlation can be carried out using additions only Also, they feature an ideal circular autocorrelation function, but in channel characterization tasks, the usual requirement is that the linear autocorrelation function be ideal, i.e., a Kronecker delta function. In this article, we consider two approaches to improving their linear autocorrelation or crosscorrelation properties: one is an inverse filtering method with theresholding and the other is a singular value decomposition (SVD) method. Both methods are simulated under noisy circumstances. The inverse filtering method resulted in an improvement in peak sidelobe level of about 1㏈ at an SNR of 30㏈, and the SVD method showed similar performances, albeit more sensitive to noise depending on the singular value selection strategy.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2002.11a
• /
• pp.407-411
• /
• 2002

Ternary sequences are digital codes consisting of discrete values -1, 0, and 1 only. They are advantageous in that the correlation can be carried out using additions only. Also, they feature an ideal circular autocorrelation function, but in channel characterization tasks, the usual requirement is that the linear autocorrelation function be ideal, i.e., a Kronecker delta function. In this article, we consider two approaches to improving their linear autocorrelation or crosscorrelation properties: one is an inverse filtering method with thresholding, and the other is a singular value decomposition (SVD) method. Both methods are simulated under noisy circumstances. The inverse filtering method resulted in an improvement in peak sidelobe level of about 11 dB at an SNR of 30 dB, and the SVD method showed similar performances, albeit more sensitive to noise depending on the singular value selection strategy.

• KIPS Transactions on Software and Data Engineering
• /
• v.2 no.8
• /
• pp.579-588
• /
• 2013

This paper proposes an effective watermarking approach for copyright protection of fire video images. The proposed watermarking approach efficiently utilizes the inherent characteristics of fire data with respect to color and texture by using a gray level co-occurrence matrix (GLCM) and fuzzy c-means (FCM) clustering. GLCM is used to generate a texture feature dataset by computing energy and homogeneity properties for each candidate fire image block. FCM is used to segment color of the fire image and to select fire texture blocks for embedding watermarks. Each selected block is then decomposed into a one-level wavelet structure with four subbands [LL, LH, HL, HH] using a discrete wavelet transform (DWT), and LH subband coefficients with a gain factor are selected for embedding watermark, where the visibility of the image does not affect. Experimental results show that the proposed watermarking approach achieves about 48 dB of high peak-signal-to-noise ratio (PSNR) and 1.6 to 2.0 of low M-singular value decomposition (M-SVD) values. In addition, the proposed approach outperforms conventional image watermarking approach in terms of normalized correlation (NC) values against several image processing attacks including noise addition, filtering, cropping, and JPEG compression.