• The Journal of the Acoustical Society of Korea
• /
• v.27 no.7
• /
• pp.386-393
• /
• 2008

This paper proposes a new approach for speech enhancement in highly nonstationary noisy environments. The spectral subtraction (SS) is a well known technique for speech enhancement in stationary noisy environments. However, in real world, noise is mostly nonstationary. The proposed method uses an auto control parameter for an adaptive threshold to work well in highly nonstationary noisy environments. Especially, the auto control parameter is affected by a linear function associated with an a posteriori signal to noise ratio (SNR) according to the increase or the decrease of the noise level. The proposed algorithm is combined with spectral subtraction (SS) using a hangover scheme (HO) for speech enhancement. The performances of the proposed method are evaluated ITU-T P.835 signal distortion (SIG) and the segment signal to-noise ratio (SNR) in various and highly nonstationary noisy environments and is superior to that of conventional spectral subtraction (SS) using a hangover (HO) and SS using a minimum statistics (MS) methods.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.6
• /
• pp.77-88
• /
• 2021

Duck industry had a rapid growth in recent years. Nevertheless, researches to improve duck house environment are still not sufficient enough. Moisture generation of duck house litter is an important factor because it may cause severe illness and low productivity. However, the measuring process is difficult because it could be disturbed with animal excrements and other factors. Therefore, it has to be calculated according to the environmental data around the duck house litter. To cut through all these procedures, we built several machine learning regression model forecasting moisture generation of litter by measured environment data (air temperature, relative humidity, wind velocity and water contents). 5 models (Multi Linear Regression, k-Nearest Neighbors, Support Vector Regression, Random Forest and Deep Neural Network). have been selected for regression. By using R-Square, RMSE and MAE as evaluation metrics, the best accurate model was estimated according to the variables for each machine learning model. In addition, to address the small amount of data acquired through lab experiments, bootstrapping method, a technique utilized in statistics, was used. As a result, the most accurate model selected was Random Forest, with parameters of n-estimator 200 by bootstrapping the original data nine times.

• Asian-Australasian Journal of Animal Sciences
• /
• v.23 no.11
• /
• pp.1445-1454
• /
• 2010

Two separate animal trials were conducted to evaluate the coincidence of dry matter intake (DMI) and average daily gain (ADG) predicted by the Cornell Net Carbohydrate and Protein System (CNCPS) and observed actually in crossbred growing bulls kept in a traditionally confined feeding system in China. In Trial 1, 45 growing Simmental${\times}$Mongolia crossbred F1 bulls were assigned to three treatments (T1-3) with 15 animals in each treatment. Trial 2 was conducted with 60 Limousin${\times}$Fuzhou crossbred F2 bulls allocated to 4 treatments (t1-4). All of the animals were confined in individual stalls. DMI and ADG for each bull were measured as a mean of each treatment. All of the data about animals, environment, management and feeds required by the CNCPS model were collected, and model predictions were generated for animals on each treatment. Subsequently, model-predicted DMI and ADG were compared with the actually recorded results. In the three treatments in Trial 1, 93.3, 80.0 and 73.3% of points fell within the range from -0.4 to 0.4 kg/d for DMI mean bias; similarly, in the four treatments in Trial 2, about 86.7, 73.3, 73.3 and 80.0% of points fell within the same range. These results indicate that the CNCPS model can accurately predict DMI of crossbred bulls in the traditionally confined feeding system in China. There were no significant differences between predicted and observed ADG for T1 (p = 0.06) and T2 (p = 0.09) in Trial 1, and for t1 (p = 0.07), t2 (p = 0.14) and t4 (p = 0.83) in Trial 2. However, significant differences between predicted and observed ADG values were observed for T3 in Trial 1 (p<0.01) and for t3 in Trial 2 (p = 0.04). By regression analysis, a statistically different value of intercept from zero for the regression equation of DMI (p<0.01) or an identical value of ADG (p = 0.06) were obtained, whereas the slopes were significantly different (p<0.01) from unity for both DMI and ADG. Additionally, small root mean square error (RMSE) values were obtained for the unbiased estimator of the two variances (DMI and ADG). Thus, the present results indicated that the CNCPS model can give acceptable estimates of DMI and ADG of crossbred growing bulls kept in a traditionally confined feeding system in China.

• The Korean Journal of Applied Statistics
• /
• v.30 no.3
• /
• pp.363-376
• /
• 2017

The insurance market is saturated and its growth engine is exhausted; consequently, the insurance industry is now in a low growth period with insurance companies that face a fierce competitive environment. In such a situation, it will be an important issue to find the probability distributions that can explain the flow of insurance claims, which are the basis of the actuarial calculation of the insurance product. Insurance claims are generally known to be well fitted by lognormal distributions or Pareto distributions biased to the left with a thick tail. In recent years, skew normal distributions or skew t distributions have been considered reasonable distributions for describing insurance claims. Cooray and Ananda (2005) proposed a composite lognormal-Pareto distribution that has the advantages of both lognormal and Pareto distributions and they also showed the composite distribution has a higher fitness than single distributions. In this paper, we introduce new composite distributions based on skew normal distributions or skew t distributions and apply them to Danish fire insurance claim data and US indemnity loss data to compare their performance with the other composite distributions and single distributions.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.32 no.10C
• /
• pp.1032-1041
• /
• 2007

The global navigation satellite system (GNSS), which is the core technique for the location based service, adopts the direct sequence/spread spectrum (DS/SS) as its modulation method. The success of a DS/SS system depends on the synchronization between the received and locally generated pseudo noise (PN) signals. As a step in the synchronization process, the tacking scheme performs fine adjustment to bring the phase difference between the two PN signals to zero. The most widely used tracking scheme is the delay locked loop with early minus late discriminator (EL-DLL). In the ideal case, the EL-DLL is the best estimator among various DLL. However, in the band-limited multipath environment, the EL-DLL has tracking bias. In this paper, the timing offset range of correlation function is divided into advanced offset range (AOR) and delayed offset range (DOR) centering around the correct synchronization time point. The tracking bias results from the following two reasons: symmetry distortion between correlation values in AOR and DOR, and mismatch between the time point corresponding to the maximum correlation value and the synchronization time point. The former and latter are named as the type I and type II tracking bias, respectively. In this paper, when the receiver has finite bandwidth in the presence of multipath signals, it is shown that the type II tracking bias becomes a more dominant error factor than the type I tracking bias, and the correlation values in AOR are not almost changed. Exploiting these characteristics, we propose a novel tracking bias mitigation scheme and demonstrate that the tracking accuracy of the proposed scheme is higher than that of the conventional scheme, both in the presence and absence of noise.

• The Journal of the Acoustical Society of Korea
• /
• v.28 no.1
• /
• pp.77-83
• /
• 2009

This paper proposes a new noise-biased compensation of minimum statistics(MS) method using a nonlinear function and a priori speech absence probability(SAP) for speech enhancement in non-stationary noisy environments. The minimum statistics(MS) method is well known technique for noise power estimation in non-stationary noisy environments. It tends to bias the noise estimate below that of true noise level. The proposed method is combined with an adaptive parameter based on a sigmoid function and a priori speech absence probability (SAP) for biased compensation. Specifically. we apply the adaptive parameter according to the a posteriori SNR. In addition, when the a priori SAP equals unity, the adaptive biased compensation factor separately increases ${\delta}_{max}$ each frequency bin, and vice versa. We evaluate the estimation of noise power capability in highly non-stationary and various noise environments, the improvement in the segmental signal-to-noise ratio (SNR), and the Itakura-Saito Distortion Measure (ISDM) integrated into a spectral subtraction (SS). The results shows that our proposed method is superior to the conventional MS approach.

• Korean Journal of Environment and Ecology
• /
• v.33 no.5
• /
• pp.605-619
• /
• 2019

The objective of this study was to analyze eutrophication characteristics, empirical model analysis, and variation of water quality according to monsoon intensity in Hapcheon Reservoir for 16 years from 2002 to 2017. Long-term annual water quality analysis showed that Hapcheon Reservoir was in a meso-nutrition to eutrophic condition, and the eutrophic state intensified after the summer monsoon. Annual rainfall volume (high vs. low rainfall) and the seasonal intensity in each year were the key factors that regulate the long-term water quality variation provided that there is no significant change of the point- and non-point source in the watershed. Dry years and wet years showed significant differences in the concentrations of TP, TN, BOD, and conductivity, indicating that precipitation had the most direct influence on nutrients and organic matter dynamics. Nutrient indicators (TP, TN), organic pollution indicators (BOD, COD), total suspended solids, and chlorophyll-a (Chl-a), which was an estimator of primary productivity, had significant positive relations (p<0.05) with precipitation. The Chl-a concentration, which is an indicator of green algae, was highly correlated with TP, TN, and BOD, which differed from other lakes that showed the lower Chl-a concentration when nutrients increased excessively. Empirical model analysis of log-transformed TN, TP, and Chl-a indicated that the Chl-a concentration was linearly regulated by phosphorus concentration, but not by nitrogen concentration. Spatial regression analysis of the riverine, transition, and lacustrine zones of $log_{10}TN$, $log_{10}TP$, and $log_{10}CHL$ showed that TN and Chl-a had significant relations (p<0.005) while TN and Chl-a had p > 0.05, indicating that phosphorus had a key role in the algal growth. Moreover, the higher correlation of both $log_{10}TP$ and $log_{10}TN$ to $log_{10}CHL$ in the riverine zone than the lacustrine zone indicated that there was little impact of inorganic suspended solids on the light limitation in the riverine zone.

• Journal of Intelligence and Information Systems
• /
• v.16 no.4
• /
• pp.67-84
• /
• 2010

In terms of business, forecasting is a work of what is expected to happen in the future to make managerial decisions and plans. Therefore, the accurate forecasting is very important for major managerial decision making and is the basis for making various strategies of business. But it is very difficult to make an unbiased and consistent estimate because of uncertainty and complexity in the future business environment. That is why we should use scientific forecasting model to support business decision making, and make an effort to minimize the model's forecasting error which is difference between observation and estimator. Nevertheless, minimizing the error is not an easy task. Case-based reasoning is a problem solving method that utilizes the past similar case to solve the current problem. To build the successful case-based reasoning models, retrieving the case not only the most similar case but also the most relevant case is very important. To retrieve the similar and relevant case from past cases, the measurement of similarities between cases is an important key factor. Especially, if the cases contain symbolic data, it is more difficult to measure the distances. The purpose of this study is to improve the forecasting accuracy of case-based reasoning approach using fuzzy relation and composition. Especially, two methods are adopted to measure the similarity between cases containing symbolic data. One is to deduct the similarity matrix following binary logic(the judgment of sameness between two symbolic data), the other is to deduct the similarity matrix following fuzzy relation and composition. This study is conducted in the following order; data gathering and preprocessing, model building and analysis, validation analysis, conclusion. First, in the progress of data gathering and preprocessing we collect data set including categorical dependent variables. Also, the data set gathered is cross-section data and independent variables of the data set include several qualitative variables expressed symbolic data. The research data consists of many financial ratios and the corresponding bond ratings of Korean companies. The ratings we employ in this study cover all bonds rated by one of the bond rating agencies in Korea. Our total sample includes 1,816 companies whose commercial papers have been rated in the period 1997~2000. Credit grades are defined as outputs and classified into 5 rating categories(A1, A2, A3, B, C) according to credit levels. Second, in the progress of model building and analysis we deduct the similarity matrix following binary logic and fuzzy composition to measure the similarity between cases containing symbolic data. In this process, the used types of fuzzy composition are max-min, max-product, max-average. And then, the analysis is carried out by case-based reasoning approach with the deducted similarity matrix. Third, in the progress of validation analysis we verify the validation of model through McNemar test based on hit ratio. Finally, we draw a conclusion from the study. As a result, the similarity measuring method using fuzzy relation and composition shows good forecasting performance compared to the similarity measuring method using binary logic for similarity measurement between two symbolic data. But the results of the analysis are not statistically significant in forecasting performance among the types of fuzzy composition. The contributions of this study are as follows. We propose another methodology that fuzzy relation and fuzzy composition could be applied for the similarity measurement between two symbolic data. That is the most important factor to build case-based reasoning model.