• 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.

• Journal of Animal Science and Technology
• /
• v.46 no.2
• /
• pp.137-144
• /
• 2004

Algorithms for estimating breeding values on several categorical data by using latent variables with threshold conception were developed and showed. Thresholds on each categorical trait were estimated by Newton’s method via gradients and Hessian matrix. This algorithm was developed by way of expansion of bivariate analysis provided by Quaas(2001). Breeding values on latent variables of categorical traits and observations on linear traits were estimated by preconditioned conjugate gradient(PCG) method, which was known having a property of fast convergence. Example was shown by simulated data with two linear traits and a categorical trait with four categories(CE=calving ease) and a dichotomous trait(SB=Still Birth) in threshold animal mixed model(TAMM). Breeding value estimates in TAMM were compared to those in linear animal mixed model (LAMM). As results, correlation estimates of breeding values to parameters were 0.91${\sim}$0.92 on CE and 0.87${\sim}$0.89 on SB in TAMM and 0.72~0.84 on CE and 0.59~0.70 on SB in LAMM. As conclusion, PCG method for estimating breeding values on several categorical traits with linear traits were feasible in TAMM.

• Journal of Cadastre & Land InformatiX
• /
• v.47 no.1
• /
• pp.37-51
• /
• 2017

Despite the improvement in accuracy of heavy rain forecasting, socioeconomic costs due to heavy rain hazards continue to increase. This is due to a lack of understanding of the effects of weather. In this study, the risk of heavy rain hazard was analyzed using the concepts of hazard, vulnerability, and exposure, which are key concepts of impact forecast presented by WMO. The potential impacts were constructed by the exposure and vulnerability variables, and the hazard index was calculated by selecting three variables according to the criteria of heavy rain warning. Weights of the potential impact index were calculated by using PCA and hazard index was calculated by applying the same weight. Correlation analysis between the potential impact index and damages showed a high correlation and it was confirmed that the potential impact index appropriately reflects the actual damage pattern. The heavy rain hazard risk was estimated by using the risk matrix consisting of the heavy rain potential impact index and the hazard index. This study provides a basis for the impacts analysis study for weather warning with spatial/temporal variation and it can be used as a useful data to establish the local heavy rain hazard prevention measures.

• The Korean Journal of Financial Management
• /
• v.13 no.1
• /
• pp.185-202
• /
• 1996

기업의 경제적 성과측정 또는 경제적 척도로서 토빈의 Q는 가치에 대해 산출된 경제적 가치를 계산함으로써 이론적인 경제적 효율성의 개념에 가장 근접한 개념으로 평가되고 있다. 그러나 정확한 계산이 어려운 토빈의 Q대신 실제로는 회계적 수익률, 주가 수익률 및 장부가격 대비 시장가치 비율(MB)등이 편의상 많이 사용되고 있다. 본 논문에서는 이러한 성과 측정 변수들간의 상관관계를 실증적으로 검증함으로써 이들 세 유형의 변수들이 토빈 Q의 대용적(代用的) 변수(變數)로서 신뢰성과 유용성이 있는지를 판정하고자 하였다. 연구 결과는 MB를 제외하고는 이들 변수들이나 또는 이들의 조합이 토빈 Q비율을 충분히 설명하기에는 너무 미흡하여 토빈 Q의 대용변수로서 경제적 효율성의 판단기준으로 사용될 경우에 그 신뢰성에 의문이 제기될 여지가 있음을 시사하고 있다. 특히, 회계적 수익률은 토빈 Q비율에 대한 설명력이 현저히 낮을 뿐 아니라, 이러한 결과는 수익률의 이동평균을 사용하거나, 연구개발비나 계열기업집단의 소속여부에 의해 잠재적인 무형자산의 영향을 반영할 수 있도록 수익률을 보완하여 사용하더라도 그 설명력이 크게 개선되지 않는 것으로 나타나고 있다.

• The Korean Journal of Applied Statistics
• /
• v.27 no.1
• /
• pp.31-41
• /
• 2014

Biplot is a useful graphical method to simultaneously explore the rows and columns of a two-way data matrix. In particular, principal component factor biplot is a graphical method to describe the interrelationship among many variables in terms of a few underlying but unobservable random variables called factors. If we consider the unobservable variables (which are mutually independent and also non-Gaussian), we can apply the independent component analysis decomposing a mixture of non-Gaussian in its independent components. In this case, if we apply the principal component factor analysis, we cannot clearly describe the interrelationship among many variables. Therefore, in this study, we apply the independent component analysis of Jutten and Herault (1991) decomposing a mixture of non-Gaussian in its independent components. We suggest an independent component biplot to interpret the independent component analysis graphically.

• Journal of Korean Society of Forest Science
• /
• v.97 no.1
• /
• pp.22-29
• /
• 2008

The objective of this study is to provide useful information for the establishment of efficient policy implementation strategies of forest road construction policy in South Korea. Data needed for the analysis was collected by a questionnaire survey. For the analysis, policy evaluation model was constructed based on theories of public policy. Evaluation model contains three independent variables (policy initiative factor, policy content, policy environment) and two dependent variables (policy result, policy impact). Since, these variables are unobservable latent variables, observable indicators are needed as proxy measures. LISREL (Linear Structural Relationships) was employed for the analysis since it is a useful measure for analysing linear structural model which consists of structural and measurement equations. It was confirmed that forest road construction is an effective policy mean for the development of rural region and activating forest resources management. The policy outcome, however, was not satisfactory. To improve the effectiveness of forest road construction policy some modification of policy contents are needed such as increased construction budget, allowing more flexibility and participation to the implementation personal and providing technical support.

• Journal of the Korean Institute of Illuminating and Electrical Installation Engineers
• /
• v.19 no.5
• /
• pp.30-38
• /
• 2005

This paper proposes a new method which can calculates nodal price as an economic signal at non-optimal operation. By using pseudo constraints in 11 cases, we calculate shadow price and nodal price based on non-optimal operation. By comparing shadow price and nodal prices based on optimal and non-optimal operation effectiveness of the method is verified.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.18 no.3
• /
• pp.425-438
• /
• 1993

Due to the producer-consumer dependency of shared variables, the potential parallelism embeded in the logic programming language has not been fully examined. The method proposed in this paper eliminates the dependency of shared variables by introducing number-sequenced variables in expanding an AND-OR proof tree. Basically, the execution of a logic program can be divided into two phases : expanding an AND-OR tree and proving the tree by matching facts with leaf nodes. In the course of the first phase, a set of number-sequenced variables are produced by expanding an AND-OR tree in the breadth-first searching. Based on the information of number-sequence, each of them is verified in the second phase in order to prove the tree. Consequently, the proposed algorithm can explore more parallelism without the dependency of shared variables.

• Korean Journal of Environmental Biology
• /
• v.41 no.4
• /
• pp.505-518
• /
• 2023

The key to invasive pest management lies in preemptive action. However, most current research using species distribution models is conducted after an invasion has occurred. This study modeled the potential distribution of the globally notorious sweet potato pest, the sweet potato weevil(Cylas formicarius), that has not yet invaded Korea using MaxEnt. Using global occurrence data, bioclimatic variables, and topsoil characteristics, MaxEnt showed high explanatory power as both the training and test areas under the curve exceeded 0.9. Among the environmental variables used in this study, minimum temperature in the coldest month (BIO06), precipitation in the driest month (BIO14), mean diurnal range (BIO02), and bulk density (BDOD) were identified as key variables. The predicted global distribution showed high values in most countries where the species is currently present, with a significant potential invasion risk in most South American countries where C. formicarius is not yet present. In Korea, Jeju Island and the southwestern coasts of Jeollanam-do showed very high probabilities. The impact of climate change under shared socioeconomic pathway (SSP) scenarios indicated an expansion along coasts as climate change progresses. By applying the 10th percentile minimum training presence rule, the potential area of occurrence was estimated at 1,439 km² under current climate conditions and could expand up to 9,485 km² under the SSP585 scenario. However, the model predicted that an inland invasion would not be serious. The results of this study suggest a need to focus on the risk of invasion in islands and coastal areas.

• The Korean Journal of Applied Statistics
• /
• v.33 no.1
• /
• pp.25-46
• /
• 2020

This paper presents ordinal probit semiparametric regression models using Bayesian Spectral Analysis Regression (BSAR) method. Ordinal probit regression is a way of modeling ordinal responses - usually more than two categories - by connecting the probability of falling into each category explained by a combination of available covariates using a probit (an inverse function of normal cumulative distribution function) link. The Bayesian probit model facilitates posterior sampling by bringing a latent variable following normal distribution, therefore, the responses are categorized by the cut-off points according to values of latent variables. In this paper, we extend the latent variable approach to a semiparametric model for the Bayesian ordinal probit regression with nonparametric functions using a spectral representation of Gaussian processes based BSAR method. The latent variable is decomposed into a parametric component and a nonparametric component with or without a shape constraint for modeling ordinal responses and predicting outcomes more flexibly. We illustrate the proposed methods with simulation studies in comparison with existing methods and real data analysis applied to a Korean National Health and Nutrition Examination Survey (KNHANES) 2016 for investigating nonparametric relationship between smoking behavior and coffee intake.