• The Korean Journal of Applied Statistics
• /
• v.5 no.1
• /
• pp.103-111
• /
• 1992

The projection pursuit method has been gussested as a technique for the analysis of the multivariate data. This method seeks out interesting linear projections of the multivariate data onto a line of a plane to solve the curse or dimensionality. In this paper we developed the discriminant analysis by using the projection method and simulations were used for comparison between this and other existing discriminant analysis methods.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.6
• /
• pp.662-667
• /
• 2022

Computing music similarity is an indispensable component in developing music search service. This paper focuses on learning a music similarity function in order to boost cover song identification performance. By using the probabilistic linear discriminant analysis, we construct a latent music space where the distances between cover song pairs reduces while the distances between the non-cover song pairs increases. We derive a music similarity function by testing hypothesis, whether two songs share the same latent variable or not, using the probabilistic models with the assumption that observed music features are generated from the learned latent music space. Experimental results performed on two cover music datasets show that the proposed music similarity improves the cover song identification performance.

• The Journal of the Acoustical Society of Korea
• /
• v.35 no.6
• /
• pp.501-509
• /
• 2016

We propose a method to improve the robustness of speaker verification on short test utterances. The accuracy of the state-of-the-art i-vector/probabilistic linear discriminant analysis systems can be degraded when testing utterance durations are short. The proposed method compensates for utterance variations of short test feature vectors using deep neural networks. We design three different types of DNN (Deep Neural Network) structures which are trained with different target output vectors. Each DNN is trained to minimize the discrepancy between the feed-forwarded output of a given short utterance feature and its original long utterance feature. We use short 2-10 s condition of the NIST (National Institute of Standards Technology, U.S.) 2008 SRE (Speaker Recognition Evaluation) corpus to evaluate the method. The experimental results show that the proposed method reduces the minimum detection cost relative to the baseline system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.10B
• /
• pp.1111-1116
• /
• 2009

In this paper, we discuss distortion-tolerant pattern recognition using computational integral imaging reconstruction. Three-dimensional object information is captured by the integral imaging pick-up process. The captured information is numerically reconstructed at arbitrary depth-levels by averaging the corresponding pixels. We apply Fisher linear discriminant analysis combined with principal component analysis to computationally reconstructed images for the distortion-tolerant recognition. Fisher linear discriminant analysis maximizes the discrimination capability between classes and principal component analysis reduces the dimensionality with the minimum mean squared errors between the original and the restored images. The presented methods provide the promising results for the classification of out-of-plane rotated objects.

• Journal of KIISE:Software and Applications
• /
• v.34 no.12
• /
• pp.1065-1074
• /
• 2007

Among various methods to compare reference signatures with an input signature, the segment-to-segment matching method has more advantages than global and point-to-point methods. However, the segment-to-segment matching method has the problem of having lower recognition rate according to the variation of partitioning points. To resolve this drawback, this paper proposes a signature verification method by considering linear discriminant analysis as well as segment-to-segment matching method. For the final decision step, we adopt statistical based Bayesian classifier technique to effectively combine two individual systems. Under the various experiments, the proposed method shows better performance than segment-to-segment based matching method.

• Journal of Korea Technology Innovation Society
• /
• v.9 no.2
• /
• pp.304-324
• /
• 2006

There were considerable researches by finance people trying to find out business ratios as predictors of corporate bankruptcy. However, such financial ratios usually lack theoretical justification to predict bankruptcy for technology-oriented small sized venture firms. This study proposes a bankruptcy predictive discriminant model using technology evaluation data instead of financial data, evaluates the model fit by the correct classification rate, cross-validation method and M-P-P method. The results indicate that linear discriminant model was found to be more appropriate model than the logistic discriminant model and 69% of original grouped data were correctly classified while 67% of future data were expected to be classified correctly.

• Journal of Welding and Joining
• /
• v.14 no.4
• /
• pp.24-32
• /
• 1996

용접공정 해석을 위한 접근방법중에서 우선적으로 결정해야할 사항으로는 비선형적인 요소와 복잡한 물리현상들을 실제적으로 해석하기위한 측정변수의 선정과 이러한 변수를 사용하여 물리적인 현상을 적절히 표현할 수 있는 알고리즘의 개발 등 을 들 수 있다. 최근까지의 연구결과를 바탕으로 해서 측정변수들의 예를 들면 용접 전류(welding current), 아크전압(arc voltage), 음향신호(acoustic signal), 아크 광(arc light) 그리고 온도(temperature)등이 있다. 용접공정을 분석하기 위한 알고 리즘으로는 확률론적 접근(statistical approach), 다양한 실험치를 이용한 인공지능 적 접근(artificial intelligence approach) 그리고 경험치를 바탕으로 인덱스(index) 을 선정하여 이를 직접 사용하는 방법 및 인공지능과 결합된 형태를 이용하는 방법등 이 있다. 또한 용접공정의 특성을 분석하기 위해서는 크게 금속이행모드(metal transfer mode), 아크의 안정성(arc stability) 그리고 용접품질(weld quality) 등을 판별할 수 있는 알고리즘의 개발이 필수적이라 할 수 있다. 본 논문에서는 용접공정 분석과 관련된 최근까지의 연구동향 및 용접신호의 특성을 좀더 심도있게 분석하기 위해 구축해야 할 필수 요건 등을 소개하고자 하며 이를 사용자가 손쉽게 이용할 수 있는 사용자 인터페이스 프로그램을 개괄적으로 설명하고자 한다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.9 no.5
• /
• pp.986-994
• /
• 2005

This Paper proposes the likelihood-based nonlinear dimension reduction method of the speech feature parameters in order to construct the voice activity detecter adaptable in noisy environment. The proposed method uses the nonlinear values of the Gaussian probability density function with the new parameters for the speec/nonspeech class. We adapted Likelihood Ratio Test to find speech part and compared its performance with that of Linear Discriminant Analysis technique. In experiments we found that the proposed method has the similar results to that of Gaussian Mixture Models.

• The Journal of the Acoustical Society of Korea
• /
• v.16 no.3
• /
• pp.99-105
• /
• 1997

This paper analyzes DLL(Delay lock loop) under the multipath fading effects. The evaluated performance measures include the steady-state timing error probability density function (PDF) and the mean-time-to-lose-lock (MTLL) under multipath fading effects. The discriminator characteristic S(${\epsilon}$) is shown to be zero at the point of timing error ${\epsilon}_{0}$ that is not zero, and the MTLL decreases as the delayed signal power $g_{2}$ and delayed time ${\tau}_{d}$ increase. We approximate the steady-state timing error PDF linearly with these variables and evaluate the steady-state timing error PDF and MTLL. The severe multipath fading effects result lower MTLL, in this case we make MTLL larger by increasing the early-late discriminator offset ${\Delta}$. First, we calculate the timing error point ${\epsilon}_{0}$, and present the performance of DLL under multipath fading. The timing error PDF, MTLL and the performance of DLL with ${\Delta}$ are also investigated. And we conclude that the larger ${\Delta}$ makes a higher MTLL and a better performance of DLL under multipath fading effects.

• Journal of the Korean Applied Science and Technology
• /
• v.39 no.3
• /
• pp.433-441
• /
• 2022

The purpose of this study is to investigate the relationship between media engagement, media empathy, and media value of MCN sports broadcasting. To achieve this purpose, a survey was conducted on 324 MCN sports broadcast viewers. Exploratory factor analysis was performed to confirm validity, and Cronbach's α test was performed to investigate reliability. In addition, correlation analysis was performed to verify discriminant validity, and linear regression analysis was performed to verify the research hypothesis, and the following conclusions were drawn. Media engagement had a positive effect on media value. Media engagement had a positive effect on media empathy. Media empathy has a positive effect on media value.