• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.358-371
• /
• 2003

Automatic word spacing is a process of deciding correct boundaries between words in a sentence including spacing errors. It is very important to increase the readability and to communicate the accurate meaning of text to the reader. The previous statistical approaches for automatic word spacing do not consider the previous spacing state, and thus can not help estimating inaccurate probabilities. In this paper, we propose two statistical word spacing models which can solve the problem of the previous statistical approaches. The proposed models are based on the observation that the automatic word spacing is regarded as a classification problem such as the POS tagging. The models can consider broader context and estimate more accurate probabilities by generalizing hidden Markov models. We have experimented the proposed models under a wide range of experimental conditions in order to compare them with the current state of the art, and also provided detailed error analysis of our models. The experimental results show that the proposed models have a syllable-unit accuracy of 98.33% and Eojeol-unit precision of 93.06% by the evaluation method considering compound nouns.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.42 no.4
• /
• pp.194-202
• /
• 2019

Reliability analysis of the components frequently starts with the data that manufacturer provides. If enough failure data are collected from the field operations, the reliability should be recomputed and updated on the basis of the field failure data. However, when the failure time record for a component contains only a few observations, all statistical methodologies are limited. In this case, where the failure records for multiple number of identical components are available, a valid alternative is combining all the data from each component into one data set with enough sample size and utilizing the useful information in the censored data. The ROK Navy has been operating multiple Patrol Killer Guided missiles (PKGs) for several years. The Korea Multi-Function Control Console (KMFCC) is one of key components in PKG combat system. The maintenance record for the KMFCC contains less than ten failure observations and a censored datum. This paper proposes a Bayesian approach with a Dirichlet mixture model to estimate failure time density for KMFCC. Trends test for each component record indicated that null hypothesis, that failure occurrence is renewal process, is not rejected. Since the KMFCCs have been functioning under different operating environment, the failure time distribution may be a composition of a number of unknown distributions, i.e. a mixture distribution, rather than a single distribution. The Dirichlet mixture model was coded as probabilistic programming in Python using PyMC3. Then Markov Chain Monte Carlo (MCMC) sampling technique employed in PyMC3 probabilistically estimated the parameters' posterior distribution through the Dirichlet mixture model. The simulation results revealed that the mixture models provide superior fits to the combined data set over single models.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.40 no.6
• /
• pp.98-105
• /
• 2003

We propose a new technique of deinterleaving multiple radar pulse sequences by means of genetic algorithm for threat identification in electronic warfare(EW) system. The conventional approaches based on histogram or continuous wavelet transform are so deterministic that they are subject to failing in detection of individual signal characteristics under real EW signal environment that suffers frequent signal missing, noise, and counter-EW signal. The proposed algorithm utilizes the probabilistic optimization procedure of genetic algorithm. This method, a time-of-arrival(TOA) only strategy, constructs an initial chromosome set using the difference of TOA. To evaluate the fitness of each gene, the defined pulse phase is considered. Since it is rare to meet with a single radar at a moment in EW field of combat, multiple solutions are to be derived in the final stage. Therefore it is designed to terminate genetic process at the prematured generation followed by a chromosome grouping. Experimental results for simulated and real radar signals show the improved performance in estimating both the number of radar and the pulse repetition interval.

• Journal of the Society of Naval Architects of Korea
• /
• v.34 no.2
• /
• pp.64-74
• /
• 1997

The necessity and importance of fatigue failure to variable load has been appreciated as the structural design technique develops and use of high tensile steel is increasing. This is much more appreciated for a large ship such as VLCC. The rigorous fatigue analysis and safety assessment should be, hence, carried out at the design stage to avoid the possibility of fatigue failure and to achieve the design result having a sufficient structural safety to fatigue strength. This paper deals with an efficient spectral fatigue analysis of ship structures by introducing the concept of stress influence coefficient. In the process included are probabilistic loading analysis, evaluation of long-term distribution of stress range and estimation of fatigue life applying the spectral fatigue analysis. An integrated computer program has been developed in which reliability analysis to fatigue strength is also included and has been applied to D/H VLCC.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.4
• /
• pp.512-517
• /
• 2011

In this paper, we propose an Active Shape Model (ASM) and Lucas-Kanade (LK) optical flow-based feature extraction and analysis method for analyzing the emotional features from facial images. Considering the facial emotion feature regions are described by Facial Action Coding System, we construct the feature-related shape models based on the combination of landmarks and extract the LK optical flow vectors at each landmarks based on the centre pixels of motion vector window. The facial emotion features are modelled by the combination of the optical flow vectors and the emotional states of facial image can be estimated by the probabilistic estimation technique, such as Bayesian classifier. Also, we extract the optimal emotional features that are considered the high correlation between feature points and emotional states by using common spatial pattern (CSP) analysis in order to improvise the operational efficiency and accuracy of emotional feature extraction process.

• Proceedings of the Korea Society of Environmental Toocicology Conference
• /
• 2003.10a
• /
• pp.91-93
• /
• 2003

It has been estimated that the equivalent of approximately $US 50 billion has been spent on research on the behavior and fate of pesticides in the environment since Rachel Carson published “Silent Spring” in 1962. Much of the resulting knowledge has been summarized explicitly in computer algorithms in a variety of empirical, deterministic, and probabilistic simulation models. These models describe and predict the transport, degradation and resultant concentrations of pesticides in various compartments of the environment during and after application. In many cases the known errors of model predictions are large. For this reason they are typically designed to be “conservative”, i.e., err on the side of over-prediction of concentrations in order to err on the side of safety. These predictions are then compared with toxicity data, from tests of the pesticide on a series of standard representative biota, including terrestrial and aquatic indicator species and higher animals (e.g., wildlife and humans). The models' predictions are good enough in some cases to provide screening of those compounds which are very unlikely to do harm, and to indicate those compounds which must be investigated further. If further investigation is indicated a more detailed (and therefore more complicated) model may be employed to give a better estimate, or field experiments may be required. A model may be used to explore “what if” questions leading to possible alternative pesticide usage patterns which give lower potential environmental concentrations and allowable exposures. We are currently at a maturing stage in this research where the knowledge base of pesticide behavior in the environmental is growing more slowly than in the past. However, innovative uses are being made of the explosion in available computer technology to use models to take ever more advantage of the knowledge we have. In this presentation, current developments in the state of the art as practiced in North America and Europe will be presented. Specifically, we will look at the efforts of the ‘Focus’ consortium in the European Union, and the ‘EMWG’ consortium in North America. These groups have been innovative in developing a process and mechanisms for discussion amongst academic, agriculture, industry and regulatory scientists, for consensus adoption of research advances into risk management methodology.

• Journal of the Korean Data and Information Science Society
• /
• v.26 no.2
• /
• pp.367-376
• /
• 2015

Today, big data has become a hot keyword in that big data may be defined as collection of data sets so huge and complex that it becomes difficult to process by traditional methods. Clustering method is to identify the information in a big database by assigning a set of objects into the clusters so that the objects in the same cluster are more similar to each other clusters. The similarity measures being used in the cluster analysis may be classified into various types depending on the nature of the data. In this paper, we computed upper and lower limits for probability interestingness measure based similarity measures without marginal probability such as Yule I and II, Michael, Digby, Baulieu, and Dispersion measure. And we compared these measures by real data and simulated experiment. By Warrens (2008), Coefficients with the same quantities in the numerator and denominator, that are bounded, and are close to each other in the ordering, are likely to be more similar. Thus, results on bounds provide means of classifying various measures. Also, knowing which coefficients are similar provides insight into the stability of a given algorithm.

• The Journal of Fisheries Business Administration
• /
• v.47 no.3
• /
• pp.71-92
• /
• 2016

This study examines the effects of consumer beliefs for food certifications on the behavioral intentions and the behavioral intention biases to purchase the certified seafoods by a subjective probability model which is on the basis of the mathematical probability model and the covariance model. The food certifications used on this study are 'Organic foods', 'Traceability system of food products' and. 'HACCP'. The representative foods of fishery products on this study is seasoned laver. The current study showed the following results. First, consumers have more than two different beliefs each for all certifications which are the subjects of this study. The beliefs of the certifications have an impact on the consumers when they consider to buy the certified seafood products. Second, consumers try to persuade by themselves to ensure that their particular belief about the certification could lead to a purchase the seafood products. Consumer beliefs of the "environmentally friendly production" on the organic foods certification is an important factor as much as the "guarantee of food safety" belief making a positive purchasing behavior intentions(PBI) bias for the organic seafood products. Consumers also have a positive PBI bias for certified seafood products in all certifications as long as a certification is considered to "guarantee the transparency of the food distribution process" as its belief. 'Traceability system' was the only one which didn't generate a positive PBI bias from the belief of "guarantee of food safety" out of three certifications.

• Journal of Digital Convergence
• /
• v.15 no.4
• /
• pp.181-190
• /
• 2017

Recently, safety accidents of daily necessities such as humidifier disinfectant, mobile phones, and infant diapers, have occurred frequently. To protect consumers from these accidents, product safety management is required, and a product risk assessment tool is needed to evaluate the degree of safety of the product. In this paper, we have constructed RAS, which is a system that can evaluate product risk based on injury information of product accident in Korea Consumer Agency. RAS consists of an injury information analysis system for analyzing accident-related information and a risk assessment system for assessing risk using information derived from the system. The Bayesian network - based probabilistic method is applied to reflect the causal relationships that affect product risk in the risk assessment process. We used RAS to evaluate 33 children's products and compared them with the results of EU RAPEX RAG. Subsequent tasks include reducing the subjectivity of the input of the accident impact scale, and linking the above two systems.

• Journal of Wetlands Research
• /
• v.11 no.1
• /
• pp.49-64
• /
• 2009

Rainfall-runoff models are used for efficient management, distribution, planning, and design of water resources in accordance with the process of hydrologic cycle. The models simplify the transition of rainfall to runoff as rainfall through different processes including evaporation, transpiration, interception, and infiltration. As the models simplify complex physical processes, gaps between the models and actual rainfall events exist. For more accurate simulation, appropriate models that suit analysis goals are selected and reliable long-term hydrological data are collected. However, uncertainty is inherent in models. It is therefore necessary to evaluate reliability of simulation results from models. A number of studies have evaluated uncertainty ingrained in rainfall-runoff models. In this paper, Meta-Gaussian method proposed by Montanari and Brath(2004) was used to assess uncertainty of simulation outputs from rainfall-runoff models. The model, which estimates upper and lower bounds of the confidence interval from probabilistic distribution of a model's error, can quantify global uncertainty of hydrological models. In this paper, Meta-Gaussian method was applied to analyze uncertainty of simulated runoff outputs from $Vflo^{TM}$, a physically-based distribution model and HEC-HMS model, a conceptual lumped model.