• MALSORI
• /
• no.57
• /
• pp.165-174
• /
• 2006

In this paper we use a training algorithm, MCE (Minimum Classification Error), to improve the performance of a text-independent speaker identification system. The MCE training scheme takes account of possible competing speaker hypotheses and tries to reduce the probability of incorrect hypotheses. Experiments performed on a small set speaker identification task show that the discriminant training method using MCE can reduce identification errors by up to 54% over a baseline system trained using Bayesian adaptation to derive GMM (Gaussian Mixture Models) speaker models from a UBM (Universal Background Model).

• Journal of Advanced Marine Engineering and Technology
• /
• v.27 no.1
• /
• pp.42-48
• /
• 2003

In general many engineers can diagnose the fault condition using the abnormal ones among data monitored from a diesel engine, but they don't need the system modelling or identification for the work. They check the abnormal data and the relationship and then catch the fault condition of the engine. This paper proposes the construction of a fault diagnosis engine through malfunction data gained from the data fault detection system of neural networks for diesel generator engine, and the rule inference method to induce the rule for fuzzy inference from the malfunction data of diesel engine like a site engineer with a fuzzy system. The proposed fault diagnosis system is constructed in the sense of the Malfunction Diagnosis Engine(MDE) and Hierarchy of Malfunction Hypotheses(HMH). The system is concerned with the rule reduction method of knowledge base for related data among the various interactive data.

• Journal of Labour Economics
• /
• v.22 no.2
• /
• pp.243-262
• /
• 1999

This paper demonstrates both theoretically and empirically that there are industry-specific factors as well as the cyclical ones that affect quit and layoff incidences and that the industry-specific effects are positively correlated between the two incidences across industries while the cyclical effects are negatively correlated over time. We first set up a theoretical model to analyze how its parameters affect quits and layoffs through the corresponding change in the optimal wage for the employer and the worker, and then derive from the theoretical separation behaviors the two testable hypotheses - that quits and layoffs are positively correlated to each other across industries, and that quits move procyclically while layoffs move countercyclically. We analyze the two sets of data, BLS establishment data on turnover rates and PSID, to empirically test and confirm each of the two hypotheses.

• Korean Management Science Review
• /
• v.27 no.3
• /
• pp.33-41
• /
• 2010

A lot of researches find negative volatility risk premium in options market. We can make a trading profit by exploiting the negative volatility premium. This study proposes negative volatility risk premium hypotheses in the KOSPI 200 stock price index options market and empirically test the proposed hypotheses with intra-day short straddle strategy. This strategy sells both at-the-money call option and at-the-money put option at market open and exits the position at market close. Using MySQL 5.1, we create our database with 1 minute option price data of the KOSPI 200 index options from 2004 to 2009. Empirical results show that negative volatility risk premium exists in the KOSPI 200 stock price index options market. Furthermore, intra-day short straddle strategy consistently produces annual profits except one year.

• Journal of the Korean Mathematical Society
• /
• v.53 no.4
• /
• pp.781-793
• /
• 2016

The solutions of equations are usually found using iterative methods whose convergence order is determined by Taylor expansions. In particular, the local convergence of the method we study in this paper is shown under hypotheses reaching the third derivative of the operator involved. These hypotheses limit the applicability of the method. In our study we show convergence of the method using only the first derivative. This way we expand the applicability of the method. Numerical examples show the applicability of our results in cases earlier results cannot.

• Ocean and Polar Research
• /
• v.25 no.1
• /
• pp.119-128
• /
• 2003

Although phoscorites and carbonatites form only a minor proportion of the earth's crustal rocks, these unusual rocks and their intimate relation are of both academic and economic importance. Rare metal (Nb, Zr, Ta) and REEs mineralizations are in close relation with the differentiation of these phoscorite-carbonatite complexes (PCCs). Recent integrated petrological and geochemical data on PCCs in the Kola Alkaline Province, Arctic, indicate that phoscorites and associated carbonatites are differentiated from common 'carbonated silicate patental magma'. Various hypotheses for the genesis of phoscorite-carbonatite complexes have been proposed during the last half-century. A simple magmatic fractionation scheme can not explain the chemical and mineralogical characteristics of phoscorite and conjugate carbonatite. Instead, the hypotheses involving liquid immiscibility and coeval accumulation processes are favored to explain the mineralogical and geochemical characteristics of phoscorite and carbonatite association.

• Communications of the Korean Mathematical Society
• /
• v.20 no.1
• /
• pp.145-159
• /
• 2005

In this paper we consider the problem of testing statistical hypotheses for unknown parameters in nonlinear regression models and propose three asymptotically equivalent tests based on regression quantiles estimators, which are Wald test, Lagrange Multiplier test and Likelihood Ratio test. We also derive the asymptotic distributions of the three test statistics both under the null hypotheses and under a sequence of local alternatives and verify that the asymptotic relative efficiency of the proposed test statistics with classical test based on least squares depends on the error distributions of the regression models. We give some examples to illustrate that the test based on the regression quantiles estimators performs better than the test based on the least squares estimators of the least absolute deviation estimators when the disturbance has asymmetric and heavy-tailed distribution.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.35 no.1
• /
• pp.154-160
• /
• 2012

This paper reviewed the two proposed hypotheses. The first hypothesis is to test the relationship between knowledge management and incremental innovation, and the second hypothesis is to test the moderating effect of firm-size on that relationship. Data were gathered from 434 employees in service industry, and multiple regression analysis was used to test two proposed hypotheses. In the first hypothesis, it was appeared that knowledge acquisition and knowledge transfer have positive effects on all incremental innovation factors (process innovation, operation innovation, service innovation). In the second hypothesis, it was appeared that firm-size has an positive interaction effect on incremental innovation in knowledge transfer, but not in knowledge acquisition. It means that knowledge transfer has more positive effect on incremental innovation in large firm-size than in small firm-size.

• Journal of the Korea Safety Management & Science
• /
• v.18 no.1
• /
• pp.131-140
• /
• 2016

This present study tested whether job stress mediates the relationship between workload and safety behavior of members in organization. In addition, we hypothesized trust in leader/coworkers functions as moderator between not only the workload-job stress link, but also the job stress-safety behavior link. In order to investigate the hypotheses, 841 employees in various fields of companies were sampled. Using structural equation modeling(SEM), we conducted moderated mediation model analysis which elaborately investigated the significance of our hypotheses. The results indicated that job stress mediated the link between workload and safety behavior of members. In addition, the relationship between workload and job stress was moderated by trust in leader/coworkers. Moreover trust in leader/coworkers moderated the job stress-safety behavior linkage. The implications and limitations of our study and suggestions for future research were discussed.

• Journal of the Korea Safety Management & Science
• /
• v.18 no.1
• /
• pp.89-97
• /
• 2016

This current research investigated whether job stress mediates the relationship between physical environment and safety behavior of members in organization. Furthermore, we hypothesized that safety climate functions as moderator between physical environment-job stress link. In order to examine the hypotheses, 870 employees in various fields of companies were sampled. Using structural equation modeling(SEM), we conducted moderated mediation model analysis which cab elaborately test the significance of our hypotheses. The results showed that job stress mediated the link between physical environment and safety behavior of members. Moreover, the relationship between physical environment and job stress was moderated by safety climate. The implications and limitations of our study and suggestions for future research were discussed.