• Speech Sciences
• /
• v.15 no.2
• /
• pp.55-65
• /
• 2008

Recently, the multi-model based speech recognizer has been used quite successfully for noisy speech recognition. For the selection of the reference HMM (hidden Markov model) which best matches the noise type and SNR (signal to noise ratio) of the input testing speech, the estimation of the SNR value using the VAD (voice activity detection) algorithm and the classification of the noise type based on the GMM (Gaussian mixture model) have been done separately in the multi-model framework. As the SNR estimation process is vulnerable to errors, we propose an efficient method which can classify simultaneously the SNR values and noise types. The KL (Kullback-Leibler) distance between the single Gaussian distributions for the noise signal during the training and testing is utilized for the classification. The recognition experiments have been done on the Aurora 2 database showing the usefulness of the model compensation method in the multi-model based speech recognizer. We could also see that further performance improvement was achievable by combining the probability density function of the MCT (multi-condition training) with that of the reference HMM compensated by the D-JA (data-driven Jacobian adaptation) in the multi-model based speech recognizer.

• Journal of Information Technology Services
• /
• v.21 no.3
• /
• pp.63-72
• /
• 2022

The extinction crisis of local cities, caused by a population density increase phenomenon in capital regions, directly causes the increase of vacant houses in local cities. According to population and housing census, Gunsan-si has continuously shown increasing trend of vacant houses during 2015 to 2019. In particular, since Gunsan-si is the city which suffers from doughnut effect and industrial decline, problems regrading to vacant house seems to exacerbate. This study aims to provide a foundation of a system which can predict and deal with the building that has high risk of becoming vacant house through implementing a data driven vacant house prediction machine learning model. Methodologically, this study analyzes three types of machine learning model by differing the data components. First model is trained based on building register, individual declared land value, house price and socioeconomic data and second model is trained with the same data as first model but with additional POI(Point of Interest) data. Finally, third model is trained with same data as the second model but with excluding water usage and electricity usage data. As a result, second model shows the best performance based on F1-score. Random Forest, Gradient Boosting Machine, XGBoost and LightGBM which are tree ensemble series, show the best performance as a whole. Additionally, the complexity of the model can be reduced through eliminating independent variables that have correlation coefficient between the variables and vacant house status lower than the 0.1 based on absolute value. Finally, this study suggests XGBoost and LightGBM based machine learning model, which can handle missing values, as final vacant house prediction model.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.6
• /
• pp.147-154
• /
• 2015

Recently agile development process became increasing effectiveness, quality, attention to experts in customer satisfaction, as evidenced in this methodology when selecting projects promoting productive efficiency. With regard to contemporary needs and user requirements on the methodology selected to meet this paper is the product based Cross functional team suggested methodology Feature Team model to solve problems of this model, and organizing the Cross functional team, this team but this outcome (product) basis, were examined for the model that points to progress the development across multiple product as a functional unit, value-driven agile project through the Skills-based model and proposed a difference. And it examined the Agile Maturity metrics. PRINCE2 Agile Health-check entries future development direction of Agile techniques is a requirement of the project outset has studied the subject objective evaluation by the assumption that they can be changed at any time, not fixed this way and for the project team through research The proposed.

• Journal of Korea Trade
• /
• v.25 no.3
• /
• pp.48-64
• /
• 2021

Purpose - This study examined the effect of tariff cuts on productivity in Korea's manufacturing industries and the effect of initial productivity level before tariff cuts on productivity improvement after tariff cuts. We also attempted to identify whether import-driven or export-driven factors are more important for productivity improvement, especially in low productivity industries. Design/methodology - Since tariff reduction is a policy decision that can affect cross-industry, its impact is spread across all industries beyond the scope of a single firm through the input and output network of industry structure. Accordingly, we proposed a new method to measure the change in productivity to reflect the impact of tariff cuts across industries. Through an Armington CGE analysis, changes in endogenous variables can be directly measured after the exogenous shock of tariff reduction, and the amount of movements in productivity triggered by tariff cuts can also be calculated. We can thus assess the effectiveness of exogenous policy, such as tariff cuts, through the difference between the benchmark and counterfactual values of endogenous variables. Findings - This study confirmed that tariff reduction positively affected productivity improvement in Korea's manufacturing industries. It also confirmed that productivity gains occur in Korea's leading export industries. Finally, greater productivity gains were recorded in the group with additional high-export-share or high-import-share conditions for low productivity industries. These results are, in a limited sense, consistent with the existing studies that emphasize the importance of exports and imports on productivity improvement, especially for low productivity industries. Originality/value - The results of our experiments are different from those of non-CGE studies, which measure the industry-level change in productivity with dummy coefficients, in terms of directly calculating the amount of change in productivity. In addition, we propose that the Armington CGE model is more appropriate than the Melitz CGE model to directly measure the productivity after tariff cuts. This is because the Melitz CGE model assumes the given specific productivity density, which does not change after an overall drop of tariffs. To the best of our knowledge, this approach to directly calculating productivity by reflecting the impact of tariff reduction across industries through CGE analysis, is unprecedented in this literature.

• Korean Journal of Air-Conditioning and Refrigeration Engineering
• /
• v.15 no.6
• /
• pp.543-550
• /
• 2003

In order to saving the energy for HVAC system of buildings, utilization of wind-induced cross ventilation is thought to be promising. However, utilization of natural ventilation alone is not sufficient for maintaining the human thermal-comfort such as in hot and humid regions. A hybrid air conditioning system with a controlled natural ventilation system, or combination of natural ventilation with mechanical air conditioning is thought to overcome the deficiency of wind-driven cross ventilation and to have significant effects on energy reduction. This paper describes a concept of hybrid system and propose a new type of hybrid system using radiational cooling with wind-induced cross ventilation. Moreover, a radiational cooling system is compared with an all-air cooling system. The characteristics of the indoor environment will be examined through CFD (Computational Fluid Dynamics) simulation, which is coupled with a radiation heat transfer simulation and with HVAC control in which the PMV value for the human model in the center of the room is controlled to attain the target value.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.22 no.5
• /
• pp.536-546
• /
• 2016

Using the inter-industry tables over the period 2010-2013 published by the Bank of Korea, this paper attempts to uncover the national economic role of the marine environmental industries through the estimation of their market size and value-added, and to analyze their economic effects through inter-industry analysis. The results show that the market size of the marine environmental industries has increased from 1.34 trillion won in 2010 to 1.97 trillion won in 2013 and their share in total national output went up from 0.04 % in 2010 to 0.05 % in 2013. Moreover, the value-added of the marine environmental industries, 618.5 billion won in 2010, amounted to 841.5 billion won in 2013 and their proportion in total national value-added has grown from 0.05 % in 2010 to 0.06 % in 2013. Three findings emerge to be used demand-driven model from the inter-industry analysis. First, the production-inducing effect of 1.0 won production or investment in the marine environmental industries has decreased from 1.8845 won in 2010 to 1.8115 won in 2013. Second, the value-added creation effect of that has declined from 0.7680 won in 2010 to 0.7063 in 2013. Third, the employment-inducing effect of 1.0 billion won production or investment in the marine environmental industries has went down from 10.17 people in 2010 to 9.18 people in 2013. In short, the market size and value-added of the marine environmental industries show an increasing trend, but their economic effects reveal a diminishing trend.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2015.05a
• /
• pp.439-439
• /
• 2015

The temporal and spatial relationship of the weather elements such as rainfall and temperature is closely linked to the streamflow simulation, especially, to the flood forecasting problems. For the study area, Imjin river basin, which has the specific characteristics in geography with river cross operation between North and South Korea, the meteorological information in the northern area is totally deficiency, lead to the inaccuracy of streamflow estimation. In the paper, this problem is solved by using the combination of global (such as soil moisture content, land use) and local hydrologic components data such as weather data (precipitation, evapotranspiration, humidity, etc.) for the model-driven runoff (surface flow, lateral flow and groundwater flow) data in each subbasin. To compute the streamflow in Imjin river basin, this study is applied the hydrologic model SURR (Sejong Univ. Rainfall-Runoff) which is the continuous rainfall-runoff model used physical foundations, originally based on Storage Function Model (SFM) to simulate the intercourse of the soil properties, weather factors and flow value. The result indicates the spatial variation in the runoff response of the different subbasins influenced by the input data. The dependancy of runoff simulation accuracy depending on the qualities of input data and model parameters is suggested in this study. The southern region with the dense of gauges and the adequate data shows the good results of the simulated discharge. Eventually, the application of SURR model in Imjin riverbasin gives the accurate consequence in simulation, and become the subsequent runoff for prediction in the future process.

• Journal of the Korean Geotechnical Society
• /
• v.22 no.6
• /
• pp.15-26
• /
• 2006

Although numerous investigations have been performed over the years to predict the behavior and bearing capacity of piles, the mechanisms are not yet entirely understood. The prediction of bearing capacity is a difficult task, because large numbers of factors affect the capacity and also have complex relationship one another. Therefore, it is extremely difficult to search the essential factors among many factors, which are related with ground condition, pile type, driving condition and others, and then appropriately consider complicated relationship among the searched factors. The present paper describes the application of Artificial Neural Network (ANN) in predicting the capacity including its components at the tip and along the shaft from dynamic load test of the driven piles. Firstly, the effect of each factor on the value of bearing capacity is investigated on the basis of sensitivity analysis using ANN modeling. Secondly, the authors use the design methodology composed of ANN and genetic algorithm (GA) to find optimal neural network model to predict the bearing capacity. The authors allow this methodology to find the appropriate combination of input parameters, the number of hidden units and the transfer structure among the input, the hidden and the out layers. The results of this study indicate that the neural network model serves as a reliable and simple predictive tool for the bearing capacity of driven piles.

• Journal of the Korean Society of Safety
• /
• v.19 no.3 s.67
• /
• pp.95-100
• /
• 2004

H-pile can be more easily driven than pipe pile by pile driver and shows high skin friction and plugging effect. And lately It is well grown that the high strength H-pile has been widely used f3r pile foundations. To compare the skin frictions of H piles under different density soil conditions, this paper presents results of a series of model tests on vertically loaded group piles. Model piles made of steel embedded in weathered granite soil were used in this study. Pile arrangements $(2\times2,\;3\tunes3)$, pile space(2D, 4D, 6D), and soil density$(D_r=40\%,\;80\%)$ were tested. The main results obtained from the model tests can be summarized as follows. The series of tests found that compression load for group piles increases as number of piles increase and piles space ratic decrease to $D_r=40\%$ of soil density. The analysis also found that the theoretical value of skin friction for group piles is greater than practical value as piles space ratio increases to $D_r=40\%$ of soil density. Piles showed the greatest difference of the skin friction in case that the pile space ratio(S/D) is 6. The theoretical value by Meyerhof and DM-7 showed 1.83 times and 1.32 times respectively as great as practical value in case of S/D=6 and $2\times2$.

• Journal of the Korean Geotechnical Society
• /
• v.21 no.3
• /
• pp.55-63
• /
• 2005

In this research, we have compared and analyzed the value driven from theoretical equation of Boussinesque, Westergaard, Newmark and K$\"{o}$ogler with our experimental value from the model test on the underground stress distribution condition. As a result of conducting the Model Test, a change in the underground stress according to the loading was proven to be very similar to the tendency shown in the theories of Westergaard. A tendency of increasing in a straight line was shown in the underground stress according to the increase of loading. When compared to that of the theoretical equation, underground stress values were great until the depth of 15cm. However, after that depth, a tendency of showing smaller value than that of the theoretical equation was shown. Correlations between Moving Repeated Load (or) and Underground Stress ($\Delta\sigma$) show $\Delta\sigma\;=\; 0.009\cdot{\sigma}r-0.1$(depth 60 cm).