• Speech Sciences
• /
• v.7 no.4
• /
• pp.11-26
• /
• 2000

In this paper, we propose the HMM with the MIN module. Because initial and re-estimated variance vectors are important elements for performance in HMM recognition systems, we propose a method which compensates for the mismatched statistical feature of training and test data. The MIN module function is a differentiable function similar to the sigmoid function. Unlike a continuous density function, it does not include variance vectors of the data set. The proposed hybrid HMM/MIN module is a unified network in which the observation probability in the HMM is replaced by the MIN module neural network. The parameters in the unified network are re-estimated by the gradient descent method for the Maximum Likelihood (ML) criterion. In estimating parameters, the variance vector is not estimated because there is no variance element in the MIN module function. The experiment was performed to compare the performance of the proposed HMM and the conventional HMM. The experiment measured an isolated number for speaker independent recognition.

• PNF and Movement
• /
• v.21 no.2
• /
• pp.185-192
• /
• 2023

Purpose: This study aimed to confirm the effectiveness of the diaphragm stretching technique as a treatment method for low back pain by evaluating maximum inspiratory pressure, maximum expiratory pressure, and changes in back mobility in patients with low back pain. Methods: Thirty-four patients with low back pain were randomly divided into two groups: an experimental group and a control group. The diaphragm stretching technique was conducted in the experimental group, and the placebo intervention was conducted in the control group. The diaphragm stretching technique was conducted once, maintaining tension for 7 min. The placebo intervention was conducted in the same position as the diaphragm stretching technique, but with only light contact maintained without pressure. Maximum inspiratory pressure, maximum expiratory pressure, and back mobility were measured before and after the intervention, and the changes were compared and analyzed. A paired sample t-test was used to compare measurements within the group before and after the intervention. An independent t-test was used to compare the experimental and control groups. Statistical significance (α) was set at 0.05. Results: In the experimental group, maximum inspiratory pressure, maximum expiratory pressure, and back mobility increased significantly after the intervention (p < 0.05). However, there was no significant difference in the changes in all areas of the control (p > 0.05). As a result of comparative analysis of changes before and after the intervention, there were significant differences in maximum inspiratory pressure, maximum expiratory pressure, and back mobility only in the experimental group (p < 0.05). Conclusion: The diaphragm stretching technique improved maximum inspiratory pressure, maximum expiratory pressure, and back mobility compared to the placebo intervention. Therefore, the diaphragm stretching technique can be recommended as a physical therapy intervention to improve pain in patients with LBP.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.1C
• /
• pp.65-71
• /
• 2004

We improved the MLLR speaker adaptation algorithm with reduction of the order of HMM parameters using PCA(Principle Component Analysis) or ICA(Independent Component Analysis). 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, minimize the correlations between data elements, and remove the axis with less covariance or higher-order statistical independencies. Ordinary MLLR algorithm needs more than 30 seconds adaptation data to represent higher word recognition rate of SD(Speaker Dependent) models than of SI(Speaker Independent) models, whereas proposed algorithm needs just more than 10 seconds adaptation data. 10 components for ICA and PCA represent similar performance with 36 components for ordinary MLLR framework. So, compared with ordinary MLLR algorithm, the amount of total computation requested in speaker adaptation is reduced by about 1/167 in proposed MLLR algorithm.

• The Transactions of the Korea Information Processing Society
• /
• v.1 no.4
• /
• pp.511-516
• /
• 1994

Parameter estimation techniques are discussed for the complex frequency analysis of an electromagnetic scatterer. The paper suggests how the Maximum Likelihood estimation technique can be applied for this purpose. Experiments on hypothetical data sets demonstrate that the Maximum Likelihood technique is better than the Pencil of Functions technique. Although there have been several techniques including MLE suggested as tools of the parameter estimation, the proposed method has strong advantages under the noise-contaminated sample data environment because it uses minimal dimension of system matrix that stands totally independent of the length of extracted data set.

• Journal of the Korean Society of Physical Medicine
• /
• v.17 no.2
• /
• pp.95-100
• /
• 2022

PURPOSE: The purpose of this study was to compare the effect of depending on the location of applying chest expansion resistance exercise on the respiratory muscle strength stroke patients, and to suggest more effective interventions to improve respiratory function in stroke patients in clinical practice. METHODS: A total of 30 subjects were selected and divided into two groups, and chest cage extension resistance exercise was applied to the sternum and rib cage, respectively, and performed for 4 weeks, 3 times a week, for 20 minutes. In order to compare the general characteristics of the study subjects and the homogeneity of the group, the pre-experimental values were analyzed using the independent sample t-test. Paired-sample t-test was used for pre-post value comparison of maximum inspiratory pressure and maximum expiratory pressure in each group. Statistical significance was set to .05. RESULTS: Both the sternum application group and the rib cage application group showed a significant difference in the maximum inspiratory pressure according to the intervention. Also, there was a statistically significant difference in the maximum expiratory pressure in the sternum application group. CONCLUSION: As breathing exercise is important for stroke patients, based on the results of this study, if therapists perform sternal extension resistance exercise or rib extension resistance exercise according to the patient's condition and environment, it can help the breathing function of stroke patients.

• Journal of the Korea Society of Computer and Information
• /
• v.20 no.1
• /
• pp.111-118
• /
• 2015

This paper deals with the routing and wavelength assignment problem (RWAP) that decides the best lightpaths for multiple packet demands for (s,t) in optical communication and assigns the minimum number of wavelengths to given lightpaths. There has been unknown of polynomial-time algorithm to obtain the optimal solution for RWAP. Hence, the RWAP is classified as NP-complete problem and one can obtain the approximate solution in polynomial-time. This paper decides the shortest main and alternate lightpath with same hop count for all (s,t) for given network in advance. When the actual demands of communication for particular multiple packet for (s,t), we decrease the maximum utilized edge into b utilized number using these dual-paths. Then, we put these (s,t) into b-wavelength bins without duplicated edge. This algorithm can be get the optimal solution within O(kn) computational complexity. For two experimental data, the proposed algorithm shows that can be obtain the known optimal solution.

• Journal of Biosystems Engineering
• /
• v.43 no.3
• /
• pp.229-236
• /
• 2018

Purpose: Makgeolli is a traditional alcoholic beverage made from rice with a fermentation starter called "nuruk." The concentration of alcohol in makgeolli depends on the temperature of the fermentation tank. It is important to monitor the alcohol concentration to manage the makgeolli production process. Methods: Data were collected from 84 makgeolli fermentation tanks over a year period. Independent variables included the temperatures of the tanks and the room where the tanks were located, as well as the quantity, acidity, and water concentration of the source. Software for the multilayer perceptron model (MLP) was written in Python using the Scikit-learn library. Results: Many models were created for which the optimization converged within 100 iterations, and their coefficients of determination $R^2$ were considerably high. The coefficient of determination $R^2$ of the best model with the training set and the test set were 0.94 and 0.93, respectively. The fact that the difference between them was very small indicated that the model was not overfitted. The maximum and minimum error was approximately 2% and the total MSE was 0.078%. Conclusions: The MLP model could help predict the alcohol concentration and to control the production process of makgeolli. In future research, the optimization of the production process will be studied based on the model.

• Journal of the Korean Society of Safety
• /
• v.32 no.1
• /
• pp.27-32
• /
• 2017

The experimental design methodology was applied in the real scale coal-fired boiler to predict the various combustion properties according to the operating conditions and to assess the coal plant safety. Response surface method (RSM) was introduced as a design of experiment, and the database for RSM was provided with the numerical simulation of the coal-fired boiler. The three independent variables, high heating value of coal (HHV), overall stoichiometry excess air ratio (OST), and burner-side stoichiometry excess air ratio (BST), were set to characterize the cross section averaged NOx concentration and temperature distribution. The maximum NOx concentration was predicted accurately and mainly controlled by BST in the boiler. The parabola function was assumed for the zone averaged peak temperature distribution, and the prediction was in a fairly good agreement with the experiments except downstream. Also, the location of the peak temperature was compared with that of maximum NOx, which implies that thermal NOx formation is the main mechanism in the coal-fired boiler. These results promise the wide use of statistical models for the fast prediction and safety assessment.

• Journal of KIISE:Computer Systems and Theory
• /
• v.26 no.8
• /
• pp.999-1007
• /
• 1999

이 논문은 재귀원형군 G(2^m , 2^k )를 그래프 이론적 관점에서 고찰하고 정점이 서로소인 사이클과 그래프 invariant에 관한 위상 특성을 제시한다. 재귀원형군은 1 에서 제안된 다중 컴퓨터의 연결망 구조이다. 재귀원형군 {{{{G(2^m , 2^k )가 길이 사이클을 가질 필요 충분 조건을 구하고, 이 조건하에서 G(2^m , 2^k )는 가능한 최대 개수의 정점이 서로소이고 길이가l`인 사이클을 가짐을 보인다. 그리고 정점 및 에지 채색, 최대 클릭, 독립 집합 및 정점 커버에 대한 그래프 invariant를 분석한다.Abstract In this paper, we investigate recursive circulant G(2^m , 2^k ) from the graph theory point of view and present topological properties of G(2^m , 2^k ) concerned with vertex-disjoint cycles and graph invariants. Recursive circulant is an interconnection structure for multicomputer networks proposed in 1 . A necessary and sufficient condition for recursive circulant {{{{G(2^m , 2^k ) to have a cycle of lengthl` is derived. Under the condition, we show that G(2^m , 2^k ) has the maximum possible number of vertex-disjoint cycles of length l`. We analyze graph invariants on vertex and edge coloring, maximum clique, independent set and vertex cover.

• Journal of Advanced Marine Engineering and Technology
• /
• v.40 no.2
• /
• pp.131-137
• /
• 2016

In reliability analysis, it is quite common for the failure of any individual or item to be attributable to more than one cause. Moreover, observed data are often censored. Recently, progressive hybrid censoring schemes have become quite popular in life-testing problems and reliability analysis. However, a limitation of the progressive hybrid censoring scheme is that it cannot be applied when few failures occur before time T. Therefore, generalized progressive hybrid censoring schemes have been introduced. In this article, we derive the likelihood inference of the unknown parameters under the assumptions that the lifetime distributions of different causes are independent and exponentially distributed. We obtain the maximum likelihood estimators of the unknown parameters in exact forms. Asymptotic confidence intervals are also proposed. Bayes estimates and credible intervals of the unknown parameters are obtained under the assumption of gamma priors on the unknown parameters. Different methods are compared using Monte Carlo simulations. One real data set is analyzed for illustrative purposes.