• Journal of the Korean Data and Information Science Society
• /
• v.18 no.1
• /
• pp.51-61
• /
• 2007

Ridge regression (RR), principal component regression (PCR) and partial least squares regression (PLS) are among popular regression methods for collinear data. While RR adds a small quantity called ridge constant to the diagonal of X'X to stabilize the matrix inversion and regression coefficients, PCR and PLS use latent variables derived from original variables to circumvent the collinearity problem. One problem of PCR and PLS is that they are very sensitive to overfitting. A new regression method is presented by combining RR and PCR and PLS, respectively, in a unified manner. It is intended to provide better predictive ability and improved stability for regression models. A real-world data from NIR spectroscopy is used to investigate the performance of the newly developed regression method.

• Communications for Statistical Applications and Methods
• /
• v.9 no.1
• /
• pp.155-166
• /
• 2002

Neural networks have been studied as a popular tool for classification and they are very flexible. Also, they are used for many applications of pattern classification and pattern recognition. This paper focuses on Bayesian approach to feed-forward neural networks with single hidden layer of units with logistic activation. In this model, we are interested in deciding the number of nodes of neural network model with p input units, one hidden layer with m hidden nodes and one output unit in Bayesian setup for fixed m. Here, we use the latent variable into the prior of the coefficient regression, and we introduce the 'sequential step' which is based on the idea of the data augmentation by Tanner and Wong(1787). The MCMC method(Gibbs sampler and Metropolish algorithm) can be used to overcome the complicated Bayesian computation. Finally, a proposed method is applied to a simulated data.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.1
• /
• pp.81-98
• /
• 2013

Since Probabilistic Latent Semantic Analysis (PLSA) and Latent Dirichlet Allocation (LDA) were introduced, many revised or extended topic models have appeared. Due to the intractable likelihood of these models, training any topic model requires to use some approximation algorithm such as variational approximation, Laplace approximation, or Markov chain Monte Carlo (MCMC). Although these approximation algorithms perform well, training a topic model is still computationally expensive given the large amount of data it requires. In this paper, we propose a new method, called non-simultaneous sampling deactivation, for efficient approximation of parameters in a topic model. While each random variable is normally sampled or obtained by a single predefined burn-in period in the traditional approximation algorithms, our new method is based on the observation that the random variable nodes in one topic model have all different periods of convergence. During the iterative approximation process, the proposed method allows each random variable node to be terminated or deactivated when it is converged. Therefore, compared to the traditional approximation ways in which usually every node is deactivated concurrently, the proposed method achieves the inference efficiency in terms of time and memory. We do not propose a new approximation algorithm, but a new process applicable to the existing approximation algorithms. Through experiments, we show the time and memory efficiency of the method, and discuss about the tradeoff between the efficiency of the approximation process and the parameter consistency.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.11
• /
• pp.375-384
• /
• 2020

This study examined the trajectories of the marital satisfaction of parents to classify its latent classes and how marital satisfaction influences the behavioral problems of their children between the identified latent classes. The 1st to 8th and 10th data from the Korea Child-Adolescent Panel Survey were analyzed using the latent class growth analysis and BCH method. First, based on the mother's trajectory of marital satisfaction, five latent classes were identified: 'low constant', 'intermediate constant', 'temporary increment-constantly decrement', 'high constant, and 'highest constant'. At the same time, based on the father's trajectory of marital satisfaction, four latent classes were identified: 'increment', 'intermediate-slightly decrement', 'high-slightly decrement', and 'highest constant'. Second, mothers with low marital satisfaction had more children with behavioral problems, and their influence had more problems with internalized behavioral problems. These problems progressed to externalized behavioral problems as they grew. Both internalized and externalized behavioral problems were also found between the identified latent classes of the father's marital satisfaction. Children of fathers with low marital satisfaction showed more behavioral problems. These findings suggest that the marital satisfaction of parents is an important variable that can influence the behavioral problems of their children.

• Journal of the Korean Regional Science Association
• /
• v.31 no.3
• /
• pp.19-37
• /
• 2015

The purpose of this study is to confirm a structural relationship on factors affecting ratio of public transportation spending to a car-owning household's total transportation expenditure. For this purpose, informations of household's attributes and activities were gathered using the 13th Korean Labor and Income Panel Study (KLIPS), and information of land-use and transportation conditions on their residential locations was collected and processed. A structural equation model (SEM) on determinants affecting ratio of public transportation expenditure was constructed, based on an execution result of factor analysis using the analyzing database. The latent variables were derived as land-use/transportation characteristic, household's attribute and household's activity. In the analyzing result of the SEM, the entire latent variables were significant. And, the first two latent variables had positive influences, and the last latent variable had a negative impact. To promote public transportation use of the car-owning households, this study suggests that the policies such as enhancement of convenience in public transportation use for the household's activities and improvement of the land-use/transport conditions are required.

• Journal of the Korea Convergence Society
• /
• v.11 no.1
• /
• pp.221-235
• /
• 2020

This study identified the latent groups according to smartphone addiction tendency and examined the factors affecting the latent group identification. The best-fitting LCA solution had three classes. The first group was 'non-academic stressed group, immersed in smartphone' It was characterized low scores on academic stress and average scores on smartphone addiction tendency. The second group was 'medium level academic stressed group, immersed in smartphone' which scored slightly above average in academic stress and smartphone addiction tendency. The third group was 'medium level academic stressed group, non-immersed in smartphone'. It showed higher scores than average in academic stress, but students with far lower scores in smartphone addiction tendency. Logistic analysis result showed that gender and grade were significant. This study is meaningful in analyzing academic related variable(academic stress) and mental health related variable(smartphone addiction tendency) to classify the groups according to patterns between the two variables and suggest appropriate intervention for each group in a convergence way.

• Management & Information Systems Review
• /
• v.38 no.1
• /
• pp.121-139
• /
• 2019

This study aims to identify the change of customer participation(CP) which is essential to the service industry as the relationship between the customer and the employee develops. The latent growth modeling analysis based on the longitudinal data is utilized to examine the pattern of the change. This is based on the fact that CP needs to be understood in the relationship and is to confirm the change in CP by the development of the relationship. Given the dynamics of the relationship, we intend to overcome the limitations of previous cross-sectional researches by revealing the trajectory of CP in the relationship through the longitudinal data. We also want to examine which variables in the relationship can facilitate changes of CP. Research has shown that CP is significantly changed with the development of the relationship when we analyzed it through latent growth modeling. This confirms that CP needs to be understood in the relationship. In addition, 'relationship proneness' variable and 'dependence to provider' variable have positive effects on the initial values of CP, but they have not been established to promote the changes of CP. Consequently, when considering the dynamics of relationships, it is important to recognize that CP is also dynamic. This study sought to get out of the cross-sectional and fragmented understanding of CP that is dynamic. Through this, we would like to propose the successful operation of the customer management program of service firms in relation to CP. This will lead to the success of service encounter where appropriate CP levels at each stage of relationship development can be achieved.

• Journal of Broadcast Engineering
• /
• v.17 no.2
• /
• pp.270-283
• /
• 2012

With the advent of multi-channel TV, IPTV and smart TV services, excessive amounts of TV program contents become available at users' sides, which makes it very difficult for TV viewers to easily find and consume their preferred TV programs. Therefore, the service of automatic TV recommendation is an important issue for TV users for future intelligent TV services, which allows to improve access to their preferred TV contents. In this paper, we present a recommendation model based on statistical machine learning using a collaborative filtering concept by taking in account both public and personal preferences on TV program contents. For this, users' preference on TV programs is modeled as a latent topic variable using LDA (Latent Dirichlet Allocation) which is recently applied in various application domains. To apply LDA for TV recommendation appropriately, TV viewers's interested topics is regarded as latent topics in LDA, and asymmetric Dirichlet distribution is applied on the LDA which can reveal the diversity of the TV viewers' interests on topics based on the analysis of the real TV usage history data. The experimental results show that the proposed LDA based TV recommendation method yields average 66.5% with top 5 ranked TV programs in weekly recommendation, average 77.9% precision in bimonthly recommendation with top 5 ranked TV programs for the TV usage history data of similar taste user groups.

• Journal of the Korean Statistical Society
• /
• v.30 no.3
• /
• pp.435-451
• /
• 2001

We describe a hierarchical bayesian model to analyze multinomial nonignorable nonresponse data. Using a Dirichlet and beta prior to model the cell probabilities, We develop a complete hierarchical bayesian analysis for multinomial proportions without making any algebraic approximation. Inference is sampling based and Markove chain Monte Carlo methods are used to perform the computations. We apply our method to the dta on body mass index(BMI) and show the model works reasonably well.

• International Journal of Reliability and Applications
• /
• v.6 no.2
• /
• pp.101-115
• /
• 2005

This article provides a new class of multivariate linear failure rate distributions where every component is a mixture of linear failure rate distribution. The new class includes several multivariate and bivariate models including Marslall and Olkin type. The approach in this paper is based on the introducing a linear failure rate distributed latent random variable. The distribution of minimum in a competing risk model is discussed.