• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.22 no.1
• /
• pp.125-130
• /
• 2018

Linear discriminant analysis is a statistical analysis method that is generally used for dimensionality reduction of the feature vectors or for class classification. However, in the case of a data set that cannot be linearly separated, it is possible to make a linear separation by mapping a feature vector into a higher dimensional space using a nonlinear function. This method is called generalized discriminant analysis or kernel discriminant analysis. In this paper, we carried out target classification experiments with active sonar target signals available on the Internet using both liner discriminant and generalized discriminant analysis methods. Experimental results are analyzed and compared with discussions. For 104 test data, LDA method has shown correct recognition rate of 73.08%, however, GDA method achieved 95.19% that is also better than the conventional MLP or kernel-based SVM.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.18 no.6
• /
• pp.1938-1944
• /
• 2004

In this study, when we make a diagnosis of constitution using QSCC Ⅱ(Questionnaire of Sasang Constitution Classification). data mining techniques are applied to seek the classification function for improving the accuracy. Data used in the analysis are the questionnaires of 1051 patients who had been treated in Dong Eui Oriental Medical Hospital and Kyung Hee Oriental Medical Hospital. The criteria for data cleansing are the response pattern in the opposite questionnaires and the positive proportion of specific questionnaires in each constitution. And the criteria for variable selection are the test of homogeneity in frequency analysis and the coefficients in the linear discriminant function. Discriminant analysis model and decision tree model are applied to seek the classification function into Sasang constitution. The accuracy in learning sample is similar in two models, the higher accuracy in test sample is obtained in discriminant analysis model.

• Journal of the Korean Data and Information Science Society
• /
• v.16 no.1
• /
• pp.19-31
• /
• 2005

Brain injured patients who had the driver's license before the injury of the brain were tested with the newly developed tool CPAD by Hangyang Medical School and the National Rehabilitation Center. The CPAD contains many variables to measure the ability of driving. Also for each patient the American standard CBDI score was measured and the result was compared with the CPAD results. Of interest is to classify the patients as pass, border, fail group after the CPAD test. To derive the discriminant functions with the group information based on CBDI, parametric/nonparametric and multivariate/univariate discriminant analysis was performed and discussed.

• Food Science of Animal Resources
• /
• v.29 no.1
• /
• pp.132-139
• /
• 2009

The objective of this study was to investigate the most effective analysis methods for palatability grading of Hanwoo beef by comparing the results of discriminant analysis with sensory data. The sensory data were obtained from sensory testing by 1,300 consumers evaluated tenderness, juiciness, flavor-likeness and overall acceptability of Hanwoo beef samples prepared by boiling, roasting and grilling cooking methods. For the discriminant analysis with one factor, overall acceptability, the linear discriminant functions and the non-parametric discriminant function with the Gaussian kernel were estimated. The linear discriminant functions were simple and easy to understand while the non-parametric discriminant functions were not explicit and had the problem of selection of kernel function and bandwidth. With the three palatability factors such as tenderness, juiciness and flavor-likeness, the canonical discriminant analysis was used and the ability of classification was calculated with the accurate classification rate and the error rate. The canonical discriminant analysis did not need the specific distributional assumptions and only used the principal component and canonical correlation. Also, it contained the function of 3 factors (tenderness, juiciness and flavor-likeness) and accurate classification rate was similar with the other discriminant methods. Therefore, the canonical discriminant analysis was the most proper method to analyze the palatability grading of Hanwoo beef.

• Journal of Korean Academy of Nursing Administration
• /
• v.13 no.4
• /
• pp.462-472
• /
• 2007

Purpose: Purpose of the this study is to define the hemodialysis noncompliance Indicators and discriminant standards levels for low Flux Hemodialysis patients and development of Hemodialysis noncompliance measurement - brief form. Method: Data was collected from 269 hemodialysis patients. To establish the hemodialysis noncompliance Indicators and to discriminate standards, 13 hemodialysis nurses and 2 nephrology doctors are participated in professional group. To verify the indicators and discriminant standards, data was ananlyzed by the canonical discriminant analysis method using by SAS 8.3 program. Result: 4 Indicators- interdialysis weight gain(IWG); average of recent 4weeks, serum phophate level, skipping of hemodialysis and hemodialysis time shortening without permission- of hemodialysis noncompliance are established and discriminant standards are developed. Discriminant ability of these 4 noncompliance indicators is 99.7%(p=.000). Hemodialysis noncompliance measurement - brief form has 96.3% discriminant accuracy. Conclusion: Hemodialysis noncompliant patients have high risks. It means that special intervention to noncompliance is needed. Also continuous and objective assessment and standards of noncompliance are needed.

• Journal of Intelligence and Information Systems
• /
• v.1 no.1
• /
• pp.91-109
• /
• 1995

This paper compared four knowledge acquisition methods (namely, neural network, case-based reasoning, discriminant analysis, and covariance structure modeling) for allergic rhinitis. The data were collected from 444 patients with suspected allergic rhinitis who visited the Otorlaryngology Deduring 1991-1993. Among four knowledge acquisition methods, the discriminant model had the best overall diagnostic capability (78%) and the neural network had slightly lower rate(76%). This may be explained by the fact that neural network is essentially non-linear discriminant model. The discriminant model was also most accurate in predicting allergic rhinitis (88%). On the other hand, the CSM had the lowest overall accuracy rate (44%) perhaps due to smaller input data set. However, it was most accuate in predicting non-allergic rhinitis (82%).

• Journal of the Korean Data and Information Science Society
• /
• v.19 no.2
• /
• pp.511-518
• /
• 2008

Medical diagnosis systems have been developed to make the knowledge and expertise of human experts more widely available, therefore achieving high-quality diagnosis. In this study, in order to support the diagnosis by the medical diagnosis system, we have preformed medical diagnosis inference three times: first by a neural network with the backpropagation algorithm, secondly by a discriminant analysis with all of the variables, and thirdly by a discriminant analysis with the selected variables. A discussion on comparison of these three methods has been provided.

• Journal of the Korean Statistical Society
• /
• v.30 no.2
• /
• pp.193-217
• /
• 2001

Despite of the rich literature in discriminant analysis, this complicated subject remains much to be explored. In this article, we study the theoretical foundation that supports Fisher's linear discriminant analysis (LDA) by setting up the classification problem under the dimension reduction framework as in Li(1991) for introducing sliced inverse regression(SIR). Through the connection between SIR and LDA, our theory helps identify sources of strength and weakness in using CRIMCOORDS(Gnanadesikan 1977) as a graphical tool for displaying group separation patterns. This connection also leads to several ways of generalizing LDA for better exploration and exploitation of nonlinear data patterns.

• Fashion & Textile Research Journal
• /
• v.8 no.1
• /
• pp.107-112
• /
• 2006

This study is done to classify the upper body shapes for late elementary schoolgirls. The sampling was done for 11~12 years-old-girls resident in Busan and Kyungnam. Based on the somatometric charateristics of them, 33 anthropometic and 7 photogrphic measurment data were acquired from every girl. These data are statistically analyzed with the following methods; Factor Analysis, Cluster Analysis, and Discriminant Analysis. Resulting from the factor analysis, it is shown that 79.95% of the whole variances can be explained with 8 factors. Through the cluster analysis, 3 types of upper body shapes can be categorized as follows: Type I has average horizontal size, big vertical size and lots of protruded chest ; Type III has big horizontal size, the mean vertical size, and big upper angle of the back ; Type II has small horizontal and vertical size and long surface length of the upper body. Through the discriminant analysis, the high discriminative items in discriminant function are follows: Upper chest circumference, arm length and waist front length of discriminant function I and waist depth, front length, back breadth, nipple to nipple breadth and upper chest circumference of discriminant function II have large coefficient values.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2129-2147
• /
• 2019

This paper introduces an adaptive face recognition method based on a Novel Incremental Kernel Nonparametric Discriminant Analysis (IKNDA) that is able to learn through time. More precisely, the IKNDA has the advantage of incrementally reducing data dimension, in a discriminative manner, as new samples are added asynchronously. Thus, it handles dynamic and large data in a better way. In order to perform face recognition effectively, we combine the Gabor features and the ordinal measures to extract the facial features that are coded across local parts, as visual primitives. The variegated ordinal measures are extraught from Gabor filtering responses. Then, the histogram of these primitives, across a variety of facial zones, is intermingled to procure a feature vector. This latter's dimension is slimmed down using PCA. Finally, the latter is treated as a facial vector input for the advanced IKNDA. A comparative evaluation of the IKNDA is performed for face recognition, besides, for other classification endeavors, in a decontextualized evaluation schemes. In such a scheme, we compare the IKNDA model to some relevant state-of-the-art incremental and batch discriminant models. Experimental results show that the IKNDA outperforms these discriminant models and is better tool to improve face recognition performance.