• 한국멀티미디어학회논문지
• /
• 제23권4호
• /
• pp.595-602
• /
• 2020

We propose a generative probabilistic model with Dirichlet prior distribution for topic modeling and text similarity analysis. It assigns a topic and calculates text correlation between documents within a corpus. It also provides posterior probabilities that are assigned to each topic of a document based on the prior distribution in the corpus. We then present a Gibbs sampling algorithm for inference about the posterior distribution and compute text correlation among 50 abstracts from the papers published by IEEE. We also conduct a supervised learning to set a benchmark that justifies the performance of the LDA (Latent Dirichlet Allocation). The experiments show that the accuracy for topic assignment to a certain document is 76% for LDA. The results for supervised learning show the accuracy of 61%, the precision of 93% and the f1-score of 96%. A discussion for experimental results indicates a thorough justification based on probabilities, distributions, evaluation metrics and correlation coefficients with respect to topic assignment.

• 디지털융복합연구
• /
• 제7권4호
• /
• pp.141-148
• /
• 2009

The purpose of this study is to examine the causal correlation between communication skill of organization and their job performance. In our model, we used these variables - innovation awareness, job satisfaction, organizational commitment as a parameter. In order to verify the hypothesis based on research model, a survey among researchers (sample size=117) who work in food company was carried out. Analyzing hypothesized model and following the statistical procedure, we could assure that there exist positive correlation with communication skill, innovation awareness, job satisfaction, organizational commitment and organizational performance. A coefficient of correlation between those variables was clearly high in the group with good communication skill. Among the parameters, the variable of innovation awareness was more effected on their job performance.

• Bulletin of the Korean Chemical Society
• /
• 제35권4호
• /
• pp.1061-1066
• /
• 2014

Time-dependent formulations of the reactive scattering theory based on the wavepacket correlation functions with the M${\phi}$ller wavepackets for the electronically nonadiabatic reactions are presented. The calculations of state-to-state reactive probabilities for the quasi-Jahn-Teller scattering model system were performed. The conical intersection (CI) effects are investigated by comparing the results of the two-surface nonadiabatic calculations and the single surface adiabatic approximation. It was found that the results of the two-surface nonadiabatic calculations show interesting features in the reaction probability due to the conical intersection. Single surface adiabatic calculations with extended Born-Oppenheimer approximation using simple wavepacket phase factor was found to be able to reproduce the CI effect semi-quantitatively, while the single surface calculations with the usual adiabatic approximation cannot describe the scattering process for the Jahn-Teller model correctly.

• Industrial Engineering and Management Systems
• /
• 제1권1호
• /
• pp.79-86
• /
• 2002

This paper proposes a new algorithm for estimating ARMA model parameters. In estimating ARMA model parameters, several methods such as generalized least square method, instrumental variable method have been developed. Among these methods, the utilization of a bootstrap type algorithm is known as one of the effective approach for the estimation, but there are cases that it does not converge. Hence, in this paper, making use of a cross correlation function and utilizing the relation of structural a priori knowledge, a new bootstrap algorithm is developed. By introducing theoretical relations, it became possible to remove terms, which is liable to include much noise. Therefore, this leads to robust parameter estimation. It is shown by numerical examples that using this algorithm, all simulation cases converge while only half cases succeeded with the previous one. As for the calculation time, judging from the fact that we got converged solutions, our proposed method is said to be superior as a whole.

• 대한원격탐사학회지
• /
• 제17권4호
• /
• pp.335-344
• /
• 2001

Texture is extensively studied in a variety of image processing applications such as image segmentation and classification because it is an important property to perceive regions and surfaces. This paper focused on the analysis and synthesis of textured single and multiband images using Markov Random Field model considering the existent spatial correlation. Especially, for multiband images, the cross-channel correlation existing between bands as well as the spatial correlation within band should be considered in the model. Although a local interaction is assumed between the specified neighboring pixels in MRF models, during the maximization process, short-term correlations among neighboring pixels develop into long-term correlations. This result in exhibiting phase transition. In this research, the role of temperature to obtain the most probable state during the sampling procedure in discrete Markov Random Fields and the stopping rule were also studied.

• Asian-Australasian Journal of Animal Sciences
• /
• 제30권8호
• /
• pp.1074-1080
• /
• 2017

Objective: The growth, carcass and retail cut yield records on 1,428 Hanwoo steers obtained through progeny testing were analyzed in this study, and their heritability and genetic relationships among the traits were estimated using animal models. Methods: Two different models were compared in this study. Each model was fitted for different fixed class effects, date of slaughter for carcass traits and batch of progeny test live measurement traits, and a choice of covariates (carcass weight in Model 1 or backfat thickness in Model 2) for carcass traits. Results: The differences in body composition among individuals were deemed being unaffected by their age at slaughter, except for carcass weight and backfat thickness. Heritability estimates of body size measurements were 0.21 to 0.36. Heritability estimates of retail cut percentage were high (0.56 from Model 1 and 0.47 from Model 2). And the heritability estimates for loin muscle percentage were 0.36 from Model 1 and 0.42 from Model 2, which were high enough to consider direct selection on carcass cutability traits as effective. The genetic correlations between body size measurements and retail cut ratio (RCR) were close to zero. But, some negative genetic correlations were found with chest girths measured at yearling (Model 1) or at 24 months of age or with chest widths. Loin muscle ratio (LMR) was genetically negatively correlated with body weights or body size measurements, in general in Model 1. These relationships were low close to zero but positive in Model 2. Phenotypic correlation between cutability traits (RCR, LMR) and live body size measurements were moderate and negative in Model 1 while those in Model 2 were all close to zero. Conclusion: Therefore, the body weights or linear body measurements at an earlier age may not be the most desirable selection traits for exploitation of correlated responses to improve loin muscle or lean meat yield.

• 한국항공우주학회지
• /
• 제39권11호
• /
• pp.1069-1076
• /
• 2011

인공위성이 궤도상에서 임무를 수행하는 동안 모든 위성 부품이 허용 온도 범위 내에서 존재하도록 하기 위하여 검증된 열모델을 개발하고, 궤도 열해석을 통하여 열적 안정성을 확보하기 위한 열설계를 수행한다. 본 연구에서는 저궤도 위성 광학탑재체의 열진공/열평형 시험 결과를 이용하여 열해석 모델을 보정하고 flight heater의 작동주기를 맞추어 줌으로써 검증된 열모델을 확보하였다. 또한 위성의 열적 안정성을 확보하기 위하여 보정이 완료된 모델을 이용하여 궤도 열해석을 수행함으로써 모든 부품이 허용온도 범위내에 존재하는 것을 확인하였다.

• 한국컴퓨터정보학회논문지
• /
• 제24권9호
• /
• pp.103-108
• /
• 2019

The purpose of this study is to identify the factors that may affect the 5-year survival of colon cancer through network model and to use it as a clinical decision supporting system for colorectal cancer patients. This study was conducted using data from 2,540 patients who underwent colorectal cancer surgery from 1996 to 2018. Eleven factors related to survival of colorectal cancer were selected by consulting medical experts and previous studies. Analysis was proceeded from the data sorted out into 1,839 patients excluding missing values and outliers. Logistic regression analysis showed that age, BMI, and heart disease were statistically significant in order to identify factors affecting 5-year survival of colorectal cancer. Additionally, a correlation analysis was carried out age, BMI, heart disease, diabetes, and other diseases were correlated with 5-year survival of colorectal cancer. Sex was related with BMI, lung disease, and liver disease. Age was associated with heart disease, heart disease, hypertension, diabetes, and other diseases, and BMI with hypertension, diabetes, and other diseases. Heart disease was associated with hypertension, diabetes, hypertension, diabetes, and other diseases. In addition, diabetes and kidney disease were associated. In the correlation analysis, the network model was constructed with the Network Correlation Coefficient less than p <0.001 as the weight. The network model showed that factors directly affecting survival were age, BMI levels, heart disease, and indirectly influencing factors were diabetes, high blood pressure, liver disease and other diseases. If the network model is used as an assistant indicator for the treatment of colorectal cancer, it could contribute to increasing the survival rate of patients.

• ETRI Journal
• /
• 제34권6호
• /
• pp.858-868
• /
• 2012

In this paper, stochastic models for the cross-correlation of multiple channels are established based on measurement data collected using a wideband multiple-input multiple-output relay Band Exploration and Channel Sounder system at 3.7 GHz. We propose models for the cross-correlation characteristics of large-scale parameters (LSPs) between two links, that is, the base station and mobile station (MS) link and the relay station and MS link. The LSPs include shadow fading, Rician K-factor, delay spread, angle spread of arrival, and angle spread of departure. Furthermore, models are established for the cross-correlation of the small-scale fading in the impulse responses of two links. The statistics of these model parameters are investigated as functions of geometrical features of the multilink. They are extracted from a large amount of cross-correlation observations, which are obtained in three measurement sites along more than one hundred measurement routes. These models can be used together with the standard single-link channel models for the generation of correlated components, for example, path clusters, in two separate channels.

• 환경위생공학
• /
• 제7권2호
• /
• pp.141-152
• /
• 1992

Simple correlation analysis, factor analysis, and multi-verite analysis have been performed to analyze the relationship between air pollution and meteorological factors for air pollution and meteorological data measured at Kwanghwamun in Seoul during the period of one year (January 1990" December 1990). As a result of simple correlation and factor analysis, $SO_2$, TSP and CO concentrations have shown high negative correlation with temperature and among these indicating that these are related with pollutants emission trend based upon heating fuel usage. Ozone has a good correlation with solar radiation and relative humidity to have a closed relation with $0_3$ generation reaction mechanism. The result of multi-variate correlation analysis shows that the concentration of $SO_2$ and CO are adequate for correlation model with ambient temperature and wind speed and 0$_{3}$ concentrations are adequate for that with solar radiation and wind speed. $SO_2$ and CO levels are considered to be affected first of all by heating fuel usage as a emission source and wind speed as a dispersion effect. The $SO_2$ concentration in the condition that the temperature fall below zero is explained by multiplicative model with wind speed, only one variable.