• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.7
• /
• pp.3594-3607
• /
• 2017

The traditional feature extraction methods such as principal component analysis (PCA) cannot obtain the local structure of the samples, and locally linear embedding (LLE) cannot obtain the global structure of the samples. However, a common drawback of existing PCA and LLE algorithm is that they cannot deal well with the sparse problem of the samples. Therefore, by integrating the globality of PCA and the locality of LLE with a sparse constraint, we developed an improved and unsupervised difference algorithm called Sparse Difference Embedding (SDE), for dimensionality reduction of high-dimensional data in small sample size problems. Significantly differing from the existing PCA and LLE algorithms, SDE seeks to find a set of perfect projections that can not only impact the locality of intraclass and maximize the globality of interclass, but can also simultaneously use the Lasso regression to obtain a sparse transformation matrix. This characteristic makes SDE more intuitive and more powerful than PCA and LLE. At last, the proposed algorithm was estimated through experiments using the Yale and AR face image databases and the USPS handwriting digital databases. The experimental results show that SDE outperforms PCA LLE and UDP attributed to its sparse discriminating characteristics, which also indicates that the SDE is an effective method for face recognition.

• Journal of Korean Ophthalmic Optics Society
• /
• v.6 no.1
• /
• pp.31-35
• /
• 2001

We have studied the auto-lensometer for several years in order to begin korean-made production of it and then developed the auto-lensmeter using with the personal computer. We introduce the most important principal of PSD device and the optical principal of measuring the power of the refraction of the lens by auto-lensometer and also explain the fabrication of the LED optical source system and PSD optical system. Finally, we found the power constant of our auto-lensmeter to be about 30 Diopter/mm.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.3
• /
• pp.317-322
• /
• 2011

Korean adults' internet addiction diagnosis measure, K-scale developed by Korea National Information Society Agency (NIA), has composed of 4 categories including 20 items. This scale can diagnose user's internet addiction with individual's questionnaire items. Most of previous research works were tried to know reasons of internet addiction and to judge whether adolescents are addicted or not with their samples. In this research, it is the goal to find the key component to judge individual's internet addiction by using a decision tree in the data mining field and a principal component analysis in statistics. From the experimental results, we would discover that tolerance and preoccupation factor is the most important one to affect adult's internet addiction.

• Journal of the Korea Society of Computer and Information
• /
• v.5 no.3
• /
• pp.46-50
• /
• 2000

Recently, Internet became a great member of people. It is increased an use of internet a population. According to internet service. Also a carrying along like PDA terminal became a small size, a light weight. A computer increased in an environment wireless computer. For the reason it s demand as necessary contents development and site construct In this thesis, mobile computer an environment is necessary an action principal and interest. wireless internet is used WAP. To recommand a WAP service and to look upon next generation.

• The Transactions of the Korean Institute of Power Electronics
• /
• v.27 no.3
• /
• pp.265-274
• /
• 2022

This paper proposes a machine learning-based screening algorithm to build the retired battery pack of the energy storage system. The proposed algorithm creates the dataset of various performance parameters of the retired battery, and this dataset is preprocessed through a principal component analysis to reduce the overfitting problem. The retried batteries with a large deviation are excluded in the dataset through a density-based spatial clustering of applications with noise, and the K-means clustering method is formulated to select the group of the retired batteries to satisfy the deviation requirement conditions. The performance of the proposed algorithm is verified based on NASA and Oxford datasets.

• International Journal of Computer Science & Network Security
• /
• v.24 no.4
• /
• pp.192-196
• /
• 2024

The memory coherence problem occurs while mapping shared virtual memory in a loosely coupled multiprocessors setup. Memory is considered coherent if a read operation provides same data written in the last write operation. The problem is addressed in the literature using different algorithms. The big question is on the correctness of such a distributed algorithm. Formal verification is the principal term for a group of techniques that routinely use an analysis that is established on mathematical transformations to conclude the rightness of hardware or software behavior in divergence to dynamic verification techniques. This paper uses UPPAAL model checker to model the dynamic distributed algorithm for shared virtual memory given by K.Li and P.Hudak. We analyse the mechanism to keep the coherence of memory in every read and write operation by using a dynamic distributed algorithm. Our results show that the dynamic distributed algorithm for shared virtual memory partially fulfils its functional requirements.

• The Journal of the Korea Contents Association
• /
• v.11 no.11
• /
• pp.185-193
• /
• 2011

There is no exact standard of detecting pulmonary tuberculosis(TB) in digital image of simple chest radiography. In this study, I experimented on the principal components analysis(PCA) algorithm in the past and suggested six other parameters as identification of TB lesions. The purpose of this study was to develop and test computer aided diagnosis(detection) method for the detection and measurement of pulmonary abnormalities on digital chest radiography. It showed comparatively low recognition diagnosis rate using PCA method, however, six kinds of texture features parameters algorithm showed similar or higher diagnosis rates of pulmonary disease than that of the clinical radiologists. Proposed algorithms using computer-aided of texture analysis can distinguish between areas of abnormality in the chest digital images, differentiate lesions having pulmonary disease. The method could be useful tool for classifying and measuring chest lesions, it would play a major role in radiologist's diagnosis of disease so as to help in pre-reading diagnosis and prevention of pulmonary tuberculosis.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.11 no.8
• /
• pp.331-337
• /
• 2001

The purpose of this study is to analyze the principal causes of vibration problem and find out the method of vibration reduction in a chip-mount system. The principal causes are investigated through measurements of vibration spectrum and model parameters. Modal parameters are obtained by using an experimental model test. Based on the model parameters from experiments. a model of finite element method is formulated. The model presents effective redesign of increasing the natural frequencies in order to reduce the vibration of a chip-mount system. Further, through computer simulation for the behavior of head to be main vibration source, the best acceleration pattern of head movement can be verified to achieve effective head-positioning and reduce the vibration due to head movement.

• 한국HCI학회:학술대회논문집
• /
• 2008.02a
• /
• pp.487-492
• /
• 2008

Facial expression recognition is an intensive research area for designing Human Computer Interfaces. In this work, we present a new facial expression recognition system utilizing Enhanced Independent Component Analysis (EICA) for feature extraction and discrete Hidden Markov Model (HMM) for recognition. Our proposed approach for the first time deals with sequential images of emotion-specific facial data analyzed with EICA and recognized with HMM. Performance of our proposed system has been compared to the conventional approaches where Principal and Independent Component Analysis are utilized for feature extraction. Our preliminary results show that our proposed algorithm produces improved recognition rates in comparison to previous works.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.2
• /
• pp.549-557
• /
• 2016

In this paper we propose multiclassification method for large data sets by ensembling least squares support vector machines (LS-SVM) with principal components instead of raw input vector. We use the revised one-vs-all method for multiclassification, which is one of voting scheme based on combining several binary classifications. The revised one-vs-all method is performed by using the hat matrix of LS-SVM ensemble, which is obtained by ensembling LS-SVMs trained using each random sample from the whole large training data. The leave-one-out cross validation (CV) function is used for the optimal values of hyper-parameters which affect the performance of multiclass LS-SVM ensemble. We present the generalized cross validation function to reduce computational burden of leave-one-out CV functions. Experimental results from real data sets are then obtained to illustrate the performance of the proposed multiclass LS-SVM ensemble.