• The Journal of the Acoustical Society of Korea
• /
• v.22 no.6
• /
• pp.452-456
• /
• 2003

We discuss how to reduce the number of inverse matrix and its dimensions requested in MLLR framework for speaker adaptation. To find a smaller set of variables with less redundancy, we adapt PCA (principal component analysis) and ICA (independent component analysis) that would give as good a representation as possible. The amount of additional computation when PCA or ICA is applied is as small as it can be disregarded. 10 components for ICA and 12 components for PCA represent similar performance with 36 components for ordinary MLLR framework. If dimension of SI model parameter is n, the amount of computation of inverse matrix in MLLR is proportioned to O(n⁴). So, compared with ordinary MLLR, the amount of total computation requested in speaker adaptation is reduced by about 1/81 in MLLR with PCA and 1/167 in MLLR with ICA.

• Analytical Science and Technology
• /
• v.24 no.6
• /
• pp.519-526
• /
• 2011

Chemometrics using near-infrared (NIR) and Raman spectroscopy have found significant uses in a variety quantitative and qualitative analyses of pharmaceutical products in complex matrixes. Most of the pharmaceutical can be measured directly with little or no sample preparation using these spectroscopic methods. During pharmaceutical manufacturing process, analytical techniques with no or less sample preparation are very critical to confirm the quality. This study showed NIR and Raman spectroscopy with principal component analysis (PCA) was very effective for the blending processing control. It is of utmost importance to evaluate critical parameters related to quality of products during pharmaceutical processing. The blending is confirmed by off-line determination of active pharmaceutical ingredient (API) by a conventional method such as high performance liquid chromatography (HPLC) and UV spectroscopy. These analytical methods are time-consuming and ineffective for real time control. This study showed the possibility for the determination of blend uniformity end-point of CR tablets with the use of both NIR and Raman spectroscopy. The samples were acquired from six positions during blending processing with U-type blender from 0 to 30 min. Using both collected NIR and Raman spectral data, principal component analysis (PCA) was used to follow the uniformity of blending and finally determine the end-point. The variation of homogeneity of six samples during blending was clearly found and blend uniformity end-point was successfully confirmed in the domains of principal component (PC) scores.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.4 s.304
• /
• pp.79-86
• /
• 2005

In this paper, we propose a new LINF(Linear Independent Non-negative Factorization) algorithm for real-time face recognition systea This system greatly consists of the two parts: 1) face extraction part; 2) face recognition part. In the face extraction Part we applied subtraction image, the detection of eye and mouth region , and normalization method, and then in the face recognition Part we used LINF in extracted face candidate region images. The existing recognition system using only PCA(Principal Component Analysis) showed low recognition rates, and it was hard in the recognition system using only LDA(Linear Discriminants Analysis) to apply LDA directly when the training set is small. To overcome these shortcomings, we reduced dimension as the matrix that had non-negative value to be different from former eigenfaces and then applied LDA to the matrix in the proposed system We have experimented using self-organized DAIJFace database and ORL database offered by AT(')T laboratory in Cambridge, U.K. to evaluate the performance of the proposed system. The experimental results showed that the proposed method outperformed PCA, LDA, ICA(Independent Component Analysis) and PLMA(PCA-based LDA mixture algorithm) method within the framework of recognition accuracy.

• ETRI Journal
• /
• v.40 no.6
• /
• pp.802-812
• /
• 2018

Disease discrimination using an electronic nose is achieved by measuring the presence of a specific gas contained in the exhaled breath of patients. Many studies have reported the presence of acetone in the breath of diabetic patients. These studies suggest that acetone can be used as a biomarker of diabetes, enabling diagnoses to be made by measuring acetone levels in exhaled breath. In this study, we perform a chemical sensor array optimization to improve the performance of an electronic nose system using Wilks' lambda, sensor selection based on a principal component (B4), and a stepwise elimination (SE) technique to detect the presence of acetone gas in human breath. By applying five different temperatures to four sensors fabricated from different synthetic materials, a total of 20 sensing combinations are created, and three sensing combinations are selected for the sensor array using optimization techniques. The measurements and analyses of the exhaled breath using the electronic nose system together with the optimized sensor array show that diabetic patients and control groups can be easily differentiated. The results are confirmed using principal component analysis (PCA).

• Korean Chemical Engineering Research
• /
• v.46 no.2
• /
• pp.383-388
• /
• 2008

The safe and efficient operation of the chemical processes has become one of the primary concerns of chemical companies, and a variety of fault diagnosis methods have been developed to diagnose faults when abnormal situations arise. Recently, many research efforts have focused on fault diagnosis methods based on quantitative history data-based methods such as statistical models. However, when the history data-based models trained with the data obtained on an operation mode are applied to another operating condition, the models can make continuous wrong diagnosis, and have limits to be applied to real chemical processes with various operation modes. In order to classify operation modes of chemical processes, this study considers three multivariate models of Euclidean distance, FDA (Fisher's Discriminant Analysis), and PCA (principal component analysis), and integrates them with process dynamics to lead dynamic Euclidean distance, dynamic FDA, and dynamic PCA. A case study of the TE (Tennessee Eastman) process having six operation modes illustrates the conclusion that dynamic PCA model shows the best classification performance.

• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.8
• /
• pp.726-732
• /
• 2016

Clustering analysis is widely used in data mining to classify data into categories on the basis of their similarity. Through the decades, many clustering techniques have been developed, including hierarchical and non-hierarchical algorithms. In gene profiling problems, because of the large number of genes and the complexity of biological networks, dimensionality reduction techniques are critical exploratory tools for clustering analysis of gene expression data. Recently, clustering analysis of applying dimensionality reduction techniques was also proposed. PCA (principal component analysis) is a popular methd of dimensionality reduction techniques for clustering problems. However, previous studies analyzed the performance of PCA for only full data sets. In this paper, to specifically and robustly evaluate the performance of PCA for clustering analysis, we exploit an improved FCBF (fast correlation-based filter) of feature selection methods for supervised clustering data sets, and employ two well-known clustering algorithms: k-means and k-medoids. Computational results from supervised data sets show that the performance of PCA is very poor for large-scale features.

• Journal of Soil and Groundwater Environment
• /
• v.22 no.3
• /
• pp.10-17
• /
• 2017

In this study, the evolution and origin of major dissolved constituents of crystalline bedrock groundwater in a coastal area were evaluated using multivariate statistical and groundwater quality analyses. The groundwater types mostly belonged to the $Na(Ca)-HCO_3$ and $Ca-HCO_3$ types, indicating the effect of cation exchange. Stable isotopes of water showed two areas divided by first and secondary evaporative effects, indicating a pattern of rapid hydrological cycling. Saturation indices of minerals showed undersaturation states. Thus, the degree of evolution of groundwater is suggested as in the low to intermediate stage, based on field and laboratory analytical conditions. According to the principal component analysis (PCA) results, the chemical components of EC, $Ca^{2+}$, $Mg^{2+}$, $K^+$, $HCO_3{^-}$, $SO{_4}^{2-}$ (PCA 1), $F^-$ (PCA 3), $SiO_2$ (PCA 4), and $Fe^{2+}$ (PCA 5) are derived from various water-rock interactions. However, $NO_3{^-}$, $Na^+$, and $Cl^-$ (PCA 2) represented the chemical characteristics of both anthropogenic sources and natural sea spray.

• Journal of the Korean Society of Food Culture
• /
• v.31 no.1
• /
• pp.105-110
• /
• 2016

The objective of this study was to determine sensory profiles of rice confections. The samples used in this study obtained from Korea (traditional Korea rice snack and local specialty rice snack) and three countries (USA, Japan, and China) were evaluated and compared. The sensory characteristics of five kinds of rice confections were evaluated using a sensory test and were analyzed via quantitative description analysis (QDA), principal component analysis (PCA), and hierarchical cluster analysis (HCA). In the descriptive analysis, 10 trained panelists evaluated sensory characteristics consisting of 19 attributes, and there were significant differences (p<0.05) among the 16 characteristics. For the descriptive data, multivariate analysis of variance was carried out and identified differences among the samples. The PCA of rice confections for the first two principal components could explain 85.66% of the variations. The Korean, Japanese, and Chinese rice confections were savory, gritty, and particle-sized, the other Korean local specialty rice confections were fruity, sweet, honey-flavored, compact, and crispy, and those from the USA were glossy, grainy, bright, adhesive, cohesive, crispy, and sweet.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.4
• /
• pp.95-102
• /
• 2012

Interactive drama is a story which requires user's free choice and participation. In this study, we grasp user's preference by making training data that utilize characters of interactive drama. Furthermore, we describe process of implementing systems which recommend new users path of stories that correspond with their preference. We used PCA and NMF to extract characteristic of preference. The success rate of recommending was 75% with PCA, while 62.5% with NMF.

• 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.