• Smart Media Journal
• /
• v.13 no.2
• /
• pp.44-51
• /
• 2024

The construction of a stress measurement detaset plays a curcial role in various modern applications. In particular, for the efficient training of artificial intelligence models for stress measurement, it is essential to compare various biases and construct a quality-controlled dataset. In this paper, we propose the construction of a stress measurement dataset with quality management through the comparison of various biases. To achieve this, we introduce strss definitions and measurement tools, the process of building an artificial intelligence stress dataset, strategies to overcome biases for quality improvement, and considerations for stress data collection. Specifically, to manage dataset quality, we discuss various biases such as selection bias, measurement bias, causal bias, confirmation bias, and artificial intelligence bias that may arise during stress data collection. Through this paper, we aim to systematically understand considerations for stress data collection and various biases that may occur during the construction of a stress dataset, contributing to the construction of a dataset with guaranteed quality by overcoming these biases.

• Communications for Statistical Applications and Methods
• /
• v.10 no.2
• /
• pp.595-601
• /
• 2003

In this paper we propose an on line training method for classification based on least squares support vector machine. Proposed method enables the computation cost to be reduced and the training to be peformed incrementally, With the incremental formulation of an inverse matrix in optimization problem, current information and new input data can be used for building the new inverse matrix for the estimation of the optimal bias and Lagrange multipliers, so the large scale matrix inversion operation can be avoided. Numerical examples are included which indicate the performance of proposed algorithm.

• Speech Sciences
• /
• v.10 no.3
• /
• pp.263-277
• /
• 2003

One of major problems in speech recognition is performance degradation due to the mismatch between the training and test environments. Recently, Stereo-based Piecewise LInear Compensation for Environments (SPLICE), which is frame-based bias removal algorithm for cepstral enhancement using stereo training data and noisy speech model as a mixture of Gaussians, was proposed and showed good performance in noisy environments. In this paper, we propose several methods to improve the conventional SPLICE. First we apply Cepstral Mean Subtraction (CMS) as a preprocessor to SPLICE, instead of applying it as a postprocessor. Secondly, to compensate residual distortion after SPLICE processing, two-stage SPLICE is proposed. Thirdly we employ phonetic information for training SPLICE model. According to experiments on the Aurora 2 database, proposed method outperformed the conventional SPLICE and we achieved a 50% decrease in word error rate over the Aurora baseline system.

• ETRI Journal
• /
• v.31 no.2
• /
• pp.121-128
• /
• 2009

In this paper, we describe a fixed-threshold sequential minimal optimization (FSMO) for structured SVM problems. FSMO is conceptually simple, easy to implement, and faster than the standard support vector machine (SVM) training algorithms for structured SVM problems. Because FSMO uses the fact that the formulation of structured SVM has no bias (that is, the threshold b is fixed at zero), FSMO breaks down the quadratic programming (QP) problems of structured SVM into a series of smallest QP problems, each involving only one variable. By involving only one variable, FSMO is advantageous in that each QP sub-problem does not need subset selection. For the various test sets, FSMO is as accurate as an existing structured SVM implementation (SVM-Struct) but is much faster on large data sets. The training time of FSMO empirically scales between O(n) and O($n^{1.2}$), while SVM-Struct scales between O($n^{1.5}$) and O($n^{1.8}$).

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.5
• /
• pp.383-389
• /
• 2004

In this paper. we propose the bias compensation methods and the eigenvoice method using the mean of dimensional eigenvoice to improve the performance of rapid speaker adaptation based on eigenvoice under mismatch between training and test environment. Experimental results for vocabulary-independent word recognition task (using PBW 452 DB) show that the proposed methods yield improvements for small adaptation data. We obtained about 22∼30% relative improvement by the bias compensation methods as amount of adaptation data varied from 1 to 50, and obtained 41% relative improvement in error rate by the eigenvoice method using the mean of dimensional eigenvoice with only single adaptation word.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2018.05a
• /
• pp.150-150
• /
• 2018

The Numerical Weather Prediction (NWP) models provide information for weather forecasts. The highly nonlinear and complex interactions in the atmosphere are simplified in meteorological models through approximations and parameterization. Therefore, the simplifications may lead to biases and errors in model results. Although the models have improved over time, the biased outputs of these models are still a matter of concern in meteorological and hydrological studies. Thus, bias removal is an essential step prior to using outputs of atmospheric models. The main idea of statistical bias correction methods is to develop a statistical relationship between modeled and observed variables over the same historical period. The Model Output Statistics (MOS) would be desirable to better match the real time forecast data with observation records. Statistical post-processing methods relate model outputs to the observed values at the sites of interest. In this study three methods are used to remove the possible biases of the real-time outputs of the Weather Research and Forecast (WRF) model in Imjin basin (North and South Korea). The post-processing techniques include the Linear Regression (LR), Linear Scaling (LS) and Power Scaling (PS) methods. The MOS techniques used in this study include three main steps: preprocessing of the historical data in training set, development of the equations, and application of the equations for the validation set. The expected results show the accuracy improvement of the real-time forecast data before and after bias correction. The comparison of the different methods will clarify the best method for the purpose of the forecast skill enhancement in a real-time case study.

• Communications for Statistical Applications and Methods
• /
• v.9 no.3
• /
• pp.809-824
• /
• 2002

we propose two methods which separate the variable selection step and the split-point selection step. We call these two algorithms as CHITES method and F&CHITES method. They adapted some of the best characteristics of CART, CHAID, and QUEST. In the first step the variable, which is most significant to predict the target class values, is selected. In the second step, the exhaustive search method is applied to find the splitting point based on the selected variable in the first step. We compared the proposed methods, CART, and QUEST in terms of variable selection bias and power, error rates, and training times. The proposed methods are not only unbiased in the null case, but also powerful for selecting correct variables in non-null cases.

• Korean Journal of Artificial Intelligence
• /
• v.5 no.1
• /
• pp.29-37
• /
• 2017

The range of problems that can be handled by the activation of big data and the development of hardware has been rapidly expanded and machine learning such as deep learning has become a very versatile technology. In this paper, mnist data set is used as experimental data, and the Cross Entropy function is used as a loss model for evaluating the efficiency of machine learning, and the value of the loss function in the steepest descent method is We applied the Gradient Descent Optimize algorithm to minimize and updated weight and bias via backpropagation. In this way we analyze optimal reliability value corresponding to the number of exercises and optimal reliability value without overfitting. And comparing the overfitting time according to the number of data changes based on the number of training times, when the training frequency was 1110 times, we obtained the result of 92%, which is the optimal reliability value without overfitting.

• The Journal of the Acoustical Society of Korea
• /
• v.20 no.2E
• /
• pp.17-24
• /
• 2001

In this paper, we propose the bootstrap and aggregating (bagging) vector quantization (VQ) classifier to improve the performance of the text-independent speaker recognition system. This method generates multiple training data sets by resampling the original training data set, constructs the corresponding VQ classifiers, and then integrates the multiple VQ classifiers into a single classifier by voting. The bagging method has been proven to greatly improve the performance of unstable classifiers. Through two different experiments, this paper shows that the VQ classifier is unstable. In one of these experiments, the bias and variance of a VQ classifier are computed with a waveform database. The variance of the VQ classifier is compared with that of the classification and regression tree (CART) classifier[1]. The variance of the VQ classifier is shown to be as large as that of the CART classifier. The other experiment involves speaker recognition. The speaker recognition rates vary significantly by the minor changes in the training data set. The speaker recognition experiments involving a closed set, text-independent and speaker identification are performed with the TIMIT database to compare the performance of the bagging VQ classifier with that of the conventional VQ classifier. The bagging VQ classifier yields improved performance over the conventional VQ classifier. It also outperforms the conventional VQ classifier in small training data set problems.

• Journal of Korean Physical Therapy Science
• /
• v.27 no.2
• /
• pp.63-79
• /
• 2020

Background: A physical therapy is required to restore motor function of stroke patients, especially gait function improvement is very important to them. Design: Systemic review. Methods: The purpose of this study is to provide objective evidence for the effectiveness of the treadmill training method applied to stroke patients. The literature search for a systematic review was performed on domestic data published from 2010 to 2020 at KISS, RISS, and DBpia. Results: 40 papers meeting the inclusion/exclusion criteria were finally selected. Data analysis was performed according to the research type, intervention method and process, measurement variables, and the risk of bias assessment was conducted as quality evaluation. Conclusion: Various training methods have been suggested and proved that treadmill intervention is an effective method for improving gait and balance. As the goal of rehabilitation for stroke patients is to improve walking and balance, we expect continuous clinical research to continue to establish a more systematic training program and present a new paradigm.