• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.6
• /
• pp.2470-2491
• /
• 2018

Web-scale open information extraction (Open IE) plays an important role in NLP tasks like acquiring common-sense knowledge, learning selectional preferences and automatic text understanding. A large number of Open IE approaches have been proposed in the last decade, and the majority of these approaches are based on supervised learning or dependency parsing. In this paper, we present a novel method for web scale open information extraction, which employs cosine distance based on Google word vector as the confidence score of the extraction. The proposed method is a purely unsupervised learning algorithm without requiring any hand-labeled training data or dependency parse features. We also present the mathematically rigorous proof for the new method with Bayes Inference and Artificial Neural Network theory. It turns out that the proposed algorithm is equivalent to Maximum Likelihood Estimation of the joint probability distribution over the elements of the candidate extraction. The proof itself also theoretically suggests a typical usage of word vector for other NLP tasks. Experiments show that the distance-based method leads to further improvements over the newly presented Open IE systems on three benchmark datasets, in terms of effectiveness and efficiency.

• Communications for Statistical Applications and Methods
• /
• v.21 no.6
• /
• pp.521-528
• /
• 2014

A stochastic Gompertz diffusion model for tumor growth is a topic of active interest as cancer is a leading cause of death in Korea. The direct maximum likelihood estimation of stochastic differential equations would be possible based on the continuous path likelihood on condition that a continuous sample path of the process is recorded over the interval. This likelihood is useful in providing a basis for the so-called continuous record or infill likelihood function and infill asymptotic. In practice, we do not have fully continuous data except a few special cases. As a result, the exact ML method is not applicable. In this paper we proposed a method of parameter estimation of stochastic Gompertz differential equation via Markov chain Monte Carlo methods that is applicable for several data structures. We compared a Markov transition data structure with a data structure that have an initial point.

• Journal of Korea Society of Industrial Information Systems
• /
• v.23 no.6
• /
• pp.79-94
• /
• 2018

The purpose of this study is to specify a probabilistic tracking mechanism for customer luxury purchase implemented by hidden Markov model, Bayesian inference, customer satisfaction and net promoter score. In this paper, we have designed a probabilistic model based on customer's actual data containing purchase or non-purchase states by tracking the SPC chain : customer satisfaction -> customer referral -> purchase/non-purchase. By applying hidden Markov model and Viterbi algorithm to marketing theory, we have developed the statistical model related to probability theories and have found the best purchase pattern scenario from customer's purchase records.

• Journal of the Korean Data and Information Science Society
• /
• v.23 no.1
• /
• pp.89-98
• /
• 2012

In this paper, the ratio of parameters in two independent Maxwell distributions is parameter of interest. We proposed test statistics, which converge to standard normal distribution, based on likelihood function. The exact distribution for testing the ratio is hard to obtain. We proposed the signed log-likelihood ratio statistic and the modified signed log-likelihood ratio statistic for testing the ratio. Through simulation, we show that the modified signed log-likelihood ratio statistic converges faster than signed log-likelihood ratio statistic to standard normal distribution. We compare two statistics in terms of type I error and power. We give an example using real data.

• Communications for Statistical Applications and Methods
• /
• v.15 no.1
• /
• pp.13-26
• /
• 2008

Inference on the present data will be more reliable when the data arising from previous similar studies are available. The data arising from previous studies are referred as historical data. The power prior is defined by the likelihood function based on the historical data to the power $a_0$, where $0\;{\le}\;a_0\;{\le}\;1$. The power prior is a useful informative prior for Bayesian inference such as model selection and model comparison. We utilize the historical data to perform multiple comparison in the one-way ANOVA model. We demonstrate our results with some simulated datasets under a simple order restriction between the treatments.

• Journal of the Korean Statistical Society
• /
• v.34 no.4
• /
• pp.331-344
• /
• 2005

Inference on current data could be more reliable if there exist similar data based on previous studies. Ibrahim and Chen (2000) utilize these data to characterize the power prior. The power prior is constructed by raising the likelihood function of the historical data to the power $a_o$, where $0\;{\le}\;a_o\;{\le}\;1$. The power prior is a useful informative prior in Bayesian inference. However, for model selection or model comparison problems, the propriety of the power prior is one of the critical issues. In this paper, we suggest two joint power priors for the power law process and show that they are proper under some conditions. We demonstrate our results with a real dataset and some simulated datasets.

• Communications for Statistical Applications and Methods
• /
• v.21 no.2
• /
• pp.193-200
• /
• 2014

It was recognized by some researchers that the disturbance variance in a censored regression model is frequently underestimated by the maximum likelihood method. This underestimation has implications for the estimation of marginal effects and asymptotic standard errors. For instance, the actual coverage probability of the confidence interval based on a maximum likelihood estimate can be significantly smaller than the nominal confidence level; consequently, a Bayesian estimation is considered to overcome this difficulty. The behaviors of the maximum likelihood and Bayesian estimators of disturbance variance are examined in a fixed effects panel regression model with a limited dependent variable, which is known to have the incidental parameter problem. Behavior under random effect assumption is also investigated.

• Journal of the Korean Data and Information Science Society
• /
• v.25 no.3
• /
• pp.633-641
• /
• 2014

The exponential distibution is one of the most popular distributions in analyzing the lifetime data. In this paper, we propose multiply Type I hybrid censoring. And this paper presents the statistical inference on the scale parameter for the exponential distribution when samples are multiply Type I hybrid censoring. The scale parameter is estimated by approximate maximum likelihood estimation methods using two different Taylor series expansion types ($AMLE_I$, $AMLE_{II}$). We also obtain the maximum likelihood estimator (MLE) of the scale parameter ${\sigma}$ under the proposed multiply Type I hybrid censored samples. We compare the estimators in the sense of the root mean square error (RMSE). The simulation procedure is repeated 10,000 times for the sample size n=20 and 40 and various censored schemes. The $AMLE_{II}$ is better than $AMLE_I$ in the sense of the RMSE.

• Communications for Statistical Applications and Methods
• /
• v.26 no.2
• /
• pp.131-148
• /
• 2019

Constant stress partially accelerated life tests are studied according to exponentiated Weibull distribution. Grounded on multiple censoring, the maximum likelihood estimators are determined in connection with unknown distribution parameters and accelerated factor. The confidence intervals of the unknown parameters and acceleration factor are constructed for large sample size. However, it is not possible to obtain the Bayes estimates in plain form, so we apply a Markov chain Monte Carlo method to deal with this issue, which permits us to create a credible interval of the associated parameters. Finally, based on constant stress partially accelerated life tests scheme with exponentiated Weibull distribution under multiple censoring, the illustrative example and the simulation results are used to investigate the maximum likelihood, and Bayesian estimates of the unknown parameters.

• Journal of the Korean Data and Information Science Society
• /
• v.10 no.1
• /
• pp.67-77
• /
• 1999

For a orientation-shift model supported on the unit sphere, Euler angles are the conventional measure to parametrize orientation-shifts. The essential role which is played by rotationally symmetry of an underlying distribution is reviewed. In this paper we propose the inference procedure based on Euler angles for the rotationally symmetric spherical distribution. The likelihood ratio test(LRT) based on the Euler angles is worked out. The asymptotic distribution of the test under the null hypotheses and certain contiguous alternatives is obtained.