• Journal of Korea Multimedia Society
• /
• v.14 no.6
• /
• pp.711-718
• /
• 2011

The k-Nearest Neighbor classifier that does not require a training phase is appropriate for a variable number of classes problem like face recognition, Recently distance metric learning methods that is trained with a given data set have reported the significant improvement of the kNN classifier. However, the performance of a distance metric learning method is variable for each application, In this paper, we focus on the face recognition and compare the performance of the state-of-the-art distance metric learning methods, Our experimental results on the public face databases demonstrate that the Mahalanobis distance metric based on PCA is still competitive with respect to both performance and time complexity in face recognition.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.6 no.1
• /
• pp.91-107
• /
• 2002

This paper presents two new self-scaling variable-metric algorithms. The first is based on a known two-parameter family of rank-two updating formulae, the second employs an initial scaling of the estimated inverse Hessian which modifies the first self-scaling algorithm. The algorithms are compared with similar published algorithms, notably those due to Oren, Shanno and Phua, Biggs and with BFGS (the best known quasi-Newton method). The best of these new and published algorithms are also modified to employ inexact line searches with marginal effect. The new algorithms are superior, especially as the problem dimension increases.

• Phonetics and Speech Sciences
• /
• v.10 no.4
• /
• pp.19-29
• /
• 2018

The present study investigated whether there are differences in articulation rate by gender, age, and individual speakers in a spontaneous speech corpus produced by 40 Seoul Korean speakers. This study measured their articulation rates using a second-per-syllable metric and a syllable-per-second metric. The findings are as follows. First, in spontaneous Seoul Korean speech, there was a gender difference in articulation rates only in age group 10-19, among whom men tended to speak faster than women. Second, individual speakers showed variability in their rates of articulation. The tendency for some speakers to speak faster than others was variable. Finally, there were metric differences in articulation rate. That is, regarding the coefficients of variation, the values of the second-per-syllable metric were much higher than those for the syllable-per-second metric. The articulation rate for the syllable-per-second metric tended to be more distinct among individual speakers. The present results imply that data gathered in a corpus of Seoul Korean spontaneous speech may reflect speaker-specific differences in articulatory movements.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1998.06a
• /
• pp.118-121
• /
• 1998

In this paper, we propose another paradigm(QNBP) to be capable of overcoming Limitations of the traditional backpropagation(SDBP). QNBPis based on the method of Quasi -Newton(variable metric) with the nomalized direction vectors and computes step size through the linear search. Simulation results showed that QNBP was definitely superior to both the stochasitc SDBP and the deterministic SDBP in terms of accuracy and rate of convergence and might sumount the problem of local minima. and there was no different between DFP+SR1 and BFGS+SR1 combined algrothms in QNBP.

• Nonlinear Functional Analysis and Applications
• /
• v.27 no.3
• /
• pp.621-640
• /
• 2022

The aim of this study is to establish some mean convergence theorems for double array of fuzzy random variables in metric space endowed with a convex combination operation under various assumptions.

• Journal of the Korean Mathematical Society
• /
• v.55 no.2
• /
• pp.447-469
• /
• 2018

Types (over parameters) in the theory of atomless random variable structures correspond precisely to (conditional) distributions in probability theory. Moreover, the logic (resp. metric) topology on the type space corresponds to the topology of weak (resp. strong) convergence of distributions. In this paper, we study metrics between types. We show that type spaces under $d^{\ast}-metric$ are isometric to Wasserstein spaces. Using optimal transport theory, two formulas for the metrics between types are given. Then, we give a new proof of an integral formula for the Wasserstein distance, and generalize some results in optimal transport theory.

• Journal of Digital Convergence
• /
• v.1 no.1
• /
• pp.45-67
• /
• 2003

Developing user success metric will make it possible to evaluate the effectiveness of different electronic government systems and will help produce empirical data relating user success to task performance in electronic government environments. The purpose of this study is that combining output-oriented and interaction-oriented variables together into a single model might explain more of the variance in the dependent variable user success in a electronic government than either type of variable could have explained alone. The hypothesized reason is that user success is the result of the reciprocal effects of the properties of the information system and several facets of psychological functioning. Thus, the complex interplay between the system and the user in certain situational contorts may produce different degrees of user success with electronic governments.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.12A
• /
• pp.1333-1339
• /
• 2004

In IS-95 and IMT-2000 systems using variable code rates and constraint lengths, this paper limits code rate 1/2 and constraint length 3 and states the effective reduction of PM(Path Metric) and BM(Branch Metric) memories and arithmetic comparative calculations with appling PVSL(Prototype Vector Selecting Logic) and LVQ(Learning Vector Quantization) in neural network to simplify systems and to decode forwardly. Regardless of extension of constraint length, this paper presents the new Vierbi decoder and the appied algorithm because new structure and algorithm can apply to the existing Viterbi decoder using only uncomplicated application and verifies the rationality of the proposed Viterbi decoder through VHDL simulation and compares the performance between the proposed Viterbi decoder and the existing.

• Journal of the Chungcheong Mathematical Society
• /
• v.26 no.2
• /
• pp.421-426
• /
• 2013

We give a series of discrete random variables which converges to a random variable whose distribution function is the Riesz-N$\acute{a}$gy-Tak$\acute{a}$cs (RNT) distribution. We show this using the correspondence theorem that if the moments coincide then their corresponding distribution functions also coincide.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.10
• /
• pp.2143-2149
• /
• 2012

DVB-S2 employs LDPC codes which approach to the Shannon's limit, since it has characteristics of a good distance, error floor does not appear. Furthermore it is possible to processes full parallel processing. However, it is very difficult to high speed decoding because of a large block size and number of many iterations. This paper present HSS algorithm to reduce the iteration numbers without performance degradation. In the flooding scheme, the decoder waits until all the check-to-variable messages are updated at all parity check nodes before computing the variable metric and updating the variable-to-check messages. The HSS algorithm is to update the variable metric on a check by check basis in the same way as one code draws benefit from the other. Eventually, LDPC decoding speed based on HSS algorithm improved 30% ~50% compared to conventional one without performance degradation.