• Journal of the Korea Society of Computer and Information
• /
• v.20 no.1
• /
• pp.21-28
• /
• 2015

Feature normalization as a pre-processing step has been widely used to reduce the effect of different scale in each feature dimension and error rate in classification. Most of the existing normalization methods, however, do not use the class labels of data points and, as a result, do not guarantee the optimality of normalization in classification aspect. A supervised rank normalization method, combination of rank normalization and supervised learning technique, was proposed and demonstrated better result than others. In this paper, another technique, training sample selection, is introduced in supervised feature normalization to reduce classification error more. Training sample selection is a common technique for increasing classification accuracy by removing noisy samples and can be applied in supervised normalization method. Two sample selection measures based on the classes of neighboring samples and the distance to neighboring samples were proposed and both of them showed better results than previous supervised rank normalization method.

• Journal of Korean Society for Quality Management
• /
• v.41 no.4
• /
• pp.709-724
• /
• 2013

Purpose: The purpose of this study is to examine five evaluation models constructed by different normalization and aggregation methods in terms of the volatility of rankings and rank reversals. We also explore how the volatility of rankings of the five models changes and how often the rank reversals occur when the outliers are removed. Methods: We used data published in the Complete University Guide 2014. Two universities with missing values were excluded from the data. The university rankings were derived by using the five models, and then each model's volatility of rankings was measured. The box-plot was used to detect outliers. Results: Model 1 has the lowest volatility among the five models whether or not the outliers are included. Model 5 has the lowest number of rank reversals. Model 3, which has been used by many institutions, appears to be in the middle among the five in terms of the volatility and the rank reversals. Conclusion: The university rankings vary from one evaluation model to another depending on what normalization and aggregation methods are used. No single model exhibits clear superiority over others in both the volatility and the rank reversal. The findings of this study are expected to provide a stepping stone toward a superior model which is both reliable and robust.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.11
• /
• pp.31-38
• /
• 2013

Feature normalization as a pre-processing step has been widely used in classification problems to reduce the effect of different scale in each feature dimension and error as a result. Most of the existing methods, however, assume some distribution function on feature distribution. Even worse, existing methods do not use the labels of data points and, as a result, do not guarantee the optimality of the normalization results in classification. In this paper, proposed is a supervised rank normalization which combines rank normalization and a supervised learning technique. The proposed method does not assume any feature distribution like rank normalization and uses class labels of nearest neighbors in classification to reduce error. SVM, in particular, tries to draw a decision boundary in the middle of class overlapping zone, the reduction of data density in that area helps SVM to find a decision boundary reducing generalized error. All the things mentioned above can be verified through experimental results.

• Genomics & Informatics
• /
• v.1 no.2
• /
• pp.94-100
• /
• 2003

Motivation: Many have observed a nonlinear relationship between the signal intensity and the transcript abundance in microarray data. The first step in analyzing the data is to normalize it properly, and this should include a correction for the nonlinearity. The commonly used linear normalization schemes do not address this problem. Results: Nonlinearity is present in both cDNA and oligonucleotide arrays, but we concentrate on the latter in this paper. Across a set of chips, we identify those genes whose within-chip ranks are relatively constant compared to other genes of similar intensity. For each gene, we compute the sum of the squares of the differences in its within-chip ranks between every pair of chips as our statistic and we select a small fraction of the genes with the minimal changes in ranks at each intensity level. These genes are most likely to be non-differentially expressed and are subsequently used in the normalization procedure. This method is a generalization of the rank-invariant normalization (Li and Wong, 2001), using all available chips rather than two at a time to gather more information, while using the chip that is least likely to be affected by nonlinear effects as the reference chip. The assumption in our method is that there are at least a small number of non­differentially expressed genes across the intensity range. The normalized expression values can be substantially different from the unnormalized values and may result in altered down-stream analysis.

• IE interfaces
• /
• v.24 no.4
• /
• pp.314-322
• /
• 2011

This study analyzes a relationship between Data Envelopment Analysis(DEA) efficiency scores and a normalization index in order to examine the validity of DEA models. A normalization index concerned in this study is 'sales per R&D project fund' which is regarded as a crucial R&D project performance evaluation index in practice. For this correlation analysis, three distinct DEA models are selected such as DEA basic model, DEA/AR-I revised model(i.e. DEA basic model with Acceptance Region Type I constraints) and Super-Efficiency(SE) model. Especially, SE model is adopted where efficient R&D projects(i.e. Decision Making Units, DMU's) with DEA efficiency score of unity from DEA basic model can be further differentiated in ranks. Considering the non-normality and outliers, two rank correlation coefficients such as Spearman's ${\rho}_s$ and Kendall's ${\tau}_B$ are investigated in addition to Pearson's ${\gamma}$. With an up-to-date empirical massive dataset of n = 482 R&D projects associated with R&D Loan Program of Korea Information Communication Promotion Fund in the year of 2011, statistically significant (+) correlations are verified between the normalization index and every model's DEA efficiency scores with all three correlation coefficients. Especially, the congruence verified in this empirical analysis can be a useful reference for enhancing the practitioner's acceptability onto DEA efficiency scores as a real-world R&D project performance evaluation index.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.29 no.5
• /
• pp.1007-1017
• /
• 2019

Currently, most biometrics system authenticates users by using single biometric information. This method has many problems such as noise problem, sensitivity to data, spoofing, a limitation of recognition rate. One method to solve this problems is to use multi biometric information. The multi biometric authentication system performs information fusion for each biometric information to generate new information, and then uses the new information to authenticate the user. Among information fusion methods, a score-level fusion method is widely used. However, there is a problem that a normalization operation is required, and even if data is same, the recognition rate varies depending on the normalization method. A rank-level fusion method that does not require normalization is proposed. However, a existing rank-level fusion methods have lower recognition rate than score-level fusion methods. To solve this problem, we propose a rank-level fusion method with higher recognition rate than a score-level fusion method using correlation coefficient. The experiment compares recognition rate of a existing rank-level fusion methods with the recognition rate of proposed method using iris information(CASIA V3) and face information(FERET V1). We also compare with score-level fusion methods. As a result, the recognition rate improve from about 0.3% to 3.3%.

• Journal of the Korean Data and Information Science Society
• /
• v.18 no.4
• /
• pp.1005-1021
• /
• 2007

Traditional procedures for DNA microarray data analysis are to preprocess and normalize the gene expression data, and then to analyze the normalized data using statistical tests. Drawbacks of the traditional methods are: genuine biological signal may be unwillingly eliminated together with artifacts, the limited number of arrays per gene make statistical tests difficult to use the normality assumption or nonparametric method, and genes are tested independently without consideration of interrelationships among genes. A novel method using average rank in each array is proposed to eliminate such drawbacks. This average rank method monitors differentially regulated genes among genetically different groups and the selected genes are somewhat different from those selected by traditional P-value method. Addition of genes selected by the average rank method to the traditional method will provide better understanding of genetic differences of groups.

• Proceedings of the KSRS Conference
• /
• 2008.03a
• /
• pp.213-218
• /
• 2008

동일 지점을 촬영한 위성영상은 위성의 센서나 영상의 취득 시기, 지형의 상태 등에 따라 그 지점에 나타나는 화소값이 일정하지 않다. 이러한 영상은 영상간 모자이크나 변화 탐지 결과에 영향을 미칠 가능성이 높으므로 방사보정(또는 방사정규화)을 통해 화소값의 차이를 최소화시킬 필요가 있다. 본 연구는 선형회귀식을 적용한 상대 방사정규화에 초점을 맞추고 있으며, 선형회귀식 구성에 필요한 PIF(Pseudo Invariant Feature)를 자동으로 추출하기 위해 Ordinal Rank 알고리즘을 적용하였다. 이 방법을 통해 각 밴드별 후보 PIF를 추출하고, 공통으로 해당되는 최종 PIF를 추출할 수 있었다. RMSE(Root Mean Square Error), Dynamic range, Coefficient of variation 등을 통해 방사보정 후의 결과를 평가해보았다. 영상회귀를 이용한 방사보정알고리즘과의 비교를 통해 제안된 알고리즘이 갖는 장점을 확인하였다.

• The Korean Journal of Nuclear Medicine Technology
• /
• v.27 no.1
• /
• pp.47-54
• /
• 2023

Purpose PET-CT imaging require an appropriate quality assurance system to achieve high efficiency and reliability. Quality control is essential for improving the quality of care and patient safety. Currently, there are performance evaluation methods of UN2-1994 and UN2-2001 proposed by NEMA and IEC for PET-CT image evaluation. In this study, we compare phantom images with the same experiments before and after PET-CT 3D normalization and well counter correction and evaluate the usefulness of quality control. Materials and methods Discovery 690 (General Electric Healthcare, USA) PET-CT equiptment was used to perform 3D normalization and well counter correction as recommended by GE Healthcare. Based on the recovery coefficients for the six spheres of the NEMA IEC Body Phantom recommended by the EARL. 20kBq/㎖ of ¹⁸F was injected into the sphere of the phantom and 2kBq/㎖ of ¹⁸F was injected into the body of phantom. PET-CT scan was performed with a radioacitivity ratio of 10:1. Images were reconstructed by appliying TOF+PSF+TOF, OSEM+PSF, OSEM and Gaussian filter 4.0, 4.5, 5.0, 5.5, 6.0, 6,5 mm with matrix size 128×128, slice thickness 3.75 mm, iteration 2, subset 16 conditions. The PET image was attenuation corrected using the CT images and analyzed using software program AW 4.7 (General Electric Healthcare, USA). The ROI was set to fit 6 spheres in the CT image, RC (Recovery Coefficient) was measured after fusion of PET and CT. Statistical analysis was performed wilcoxon signed rank test using R. Results Overall, after the quality control items were performed, the recovery coefficient of the phantom image increased and measured. Recovery coefficient according to the image reconstruction increased in the order TOF+PSF, TOF, OSEM+PSF, before and after quality control, RCmax increased by OSEM 0.13, OSEM+PSF 0.16, TOF 0.16, TOF+PSF 0.15 and RCmean increased by OSEM 0.09, OSEM+PSF 0.09, TOF 0.106, TOF+PSF 0.10. Both groups showed a statistically significant difference in Wilcoxon signed rank test results (P value<0.001). Conclusion PET-CT system require quality assurance to achieve high efficiency and reliability. Standardized intervals and procedures should be followed for quality control. We hope that this study will be a good opportunity to think about the importance of quality control in PET-CT

• Journal of Information Technology Applications and Management
• /
• v.25 no.1
• /
• pp.67-85
• /
• 2018

We propose a noise reduced risk aversion index for measuring risk aversion through a laboratory experiment to overcome disadvantages of the multiple pricing list format developed by Holt and Laury (2002). We use randomized multiple list choices with coarser classification and reward weighting, supplement the rank of risk aversion with extra individual characteristics of risk attitude, and construct an index of risk aversion by standardizing the risk aversion ranking with quantile normalization. Our method reduces multiple switching problems that noisy decision makers mistakenly commit in experimental approaches, so that it is free of the framing effect which severely occurred in the HL. Furthermore, the index doesn't utilize any specific utility function or probability weighting, which allows researcher to hold the independence axiom. Since our noise reduced index of risk aversion has many good traits, it is widely used and applied to reveal fundamental characteristics of risk-related behaviors in economics and finance regardless of experimental environment.