• The Journal of the Acoustical Society of Korea
• /
• v.35 no.2
• /
• pp.134-142
• /
• 2016

In this paper, a method synthesizing speech signal using the 40 kHz ultrasonic signals reflected from the articulatory muscles was introduced and performance was evaluated. When the ultrasound signals are radiated to articulating face, the Doppler effects caused by movements of lips, jaw, and chin observed. The signals that have different frequencies from that of the transmitted signals are found in the received signals. These ADS (Acoustic-Doppler Signals) were used for estimating of the speech parameters in this study. Prior to synthesizing speech signal, a quantitative correlation analysis between ADS and speech signals was carried out on each frequency bin. According to the results, the feasibility of the ADS-based speech synthesis was validated. ADS-to-speech transformation was achieved by the joint Gaussian mixture model-based conversion rules. The experimental results from the 5 subjects showed that filter bank energy and LPC (Linear Predictive Coefficient) cepstrum coefficients are the optimal features for ADS, and speech, respectively. In the subjective evaluation where synthesized speech signals were obtained using the excitation sources extracted from original speech signals, it was confirmed that the ADS-to-speech conversion method yielded 72.2 % average recognition rates.

• Journal of Navigation and Port Research
• /
• v.45 no.3
• /
• pp.165-172
• /
• 2021

Breaking wave is one of the important design factors in the design of coastal and port structures as they are directly related to various physical phenomena occurring on the coast, such as onshore currents, sediment transport, shock wave pressure, and energy dissipation. Due to the inherent complexity of the breaking wave, many empirical formulas have been proposed to predict breaker indices such as wave breaking height and breaking depth using hydraulic models. However, the existing empirical equations for breaker indices mainly were proposed via statistical analysis of experimental data under the assumption of a specific equation. In this study, a new Munk-type empirical equation was proposed to predict the height of breaking waves based on a representative linear supervised machine learning technique with high predictive performance in various research fields related to regression or classification challenges. Although the newly proposed breaker height formula was a simple polynomial equation, its predictive performance was comparable to that of the currently available empirical formula.

• Journal of radiological science and technology
• /
• v.46 no.2
• /
• pp.131-139
• /
• 2023

High-dose I-131 used for the treatment of thyroid cancer causes localized exposure among radiology technologists handling it. There is a delay between the calibration date and when the dose of I-131 is administered to a patient. Therefore, it is necessary to directly measure the radioactivity of the administered dose using a dose calibrator. In this study, we attempted to apply machine learning modeling to measured external dose rates from shielded I-131 in order to predict their radioactivity. External dose rates were measured at 1 m, 0.3 m, and 0.1 m distances from a shielded container with the I-131, with a total of 868 sets of measurements taken. For the modeling process, we utilized the hold-out method to partition the data with a 7:3 ratio (609 for the training set:259 for the test set). For the machine learning algorithms, we chose linear regression, decision tree, random forest and XGBoost. To evaluate the models, we calculated root mean square error (RMSE), mean square error (MSE), and mean absolute error (MAE) to evaluate accuracy and R² to evaluate explanatory power. Evaluation results are as follows. Linear regression (RMSE 268.15, MSE 71901.87, MAE 231.68, R² 0.92), decision tree (RMSE 108.89, MSE 11856.92, MAE 19.24, R² 0.99), random forest (RMSE 8.89, MSE 79.10, MAE 6.55, R² 0.99), XGBoost (RMSE 10.21, MSE 104.22, MAE 7.68, R² 0.99). The random forest model achieved the highest predictive ability. Improving the model's performance in the future is expected to contribute to lowering exposure among radiology technologists.

• Bulletin of the Korean Chemical Society
• /
• v.27 no.2
• /
• pp.211-218
• /
• 2006

This work deals with 24 substances composed of nitrogen-containing heterocycles. The relationships between the chromatographic retention factor (k) and those physicochemical properties which are relevant in quantitative structure-properties relationship (QSPR) studies, such as the polarizability $(\alpha)$, molar refractivity (MR), lipophilicity (logP), dipole moment $(\mu)$, total energy $(E_{tot})$, heat of formation $(\Delta H_f)$, molecular surface area $(S_M)$, and binding energy $(E_b)$, were investigated. The accuracy of the simple linear regressions between the chromatographic retention and the descriptors for all of the compounds was satisfactory (correlation coefficient, $0.8 \leq r \leq 1.0$). The QSPR models of these nitrogen-containing heterocyclic compounds could be predicted with a multiple linear regression equation having the statistical index, r = 1.000. This work demonstrated the successful application of the multiple linear approaches through the development of accurate predictive equations for retention factors in liquid chromatography.

• Bulletin of the Korean Chemical Society
• /
• v.26 no.1
• /
• pp.139-145
• /
• 2005

Artificial neural networks (ANNs), for a first time, were successfully developed for the modeling and prediction of solvent effects on rate constant of [2+2] cycloaddition reaction of diethyl azodicarboxylate with ethyl vinyl ether in various solvents with diverse chemical structures using quantitative structure-activity relationship. The most positive charge of hydrogen atom (q$^+$), dipole moment ($\mu$), the Hildebrand solubility parameter (${\delta}_H^2$) and total charges in molecule (q$_t$) are inputs and output of ANN is log k$_2$ . For evaluation of the predictive power of the generated ANN, the optimized network with 68 various solvents as training set was used to predict log k$_2$ of the reaction in 16 solvents in the prediction set. The results obtained using ANN was compared with the experimental values as well as with those obtained using multi-parameter linear regression (MLR) model and showed superiority of the ANN model over the regression model. Mean square error (MSE) of 0.0806 for the prediction set by MLR model should be compared with the value of 0.0275 for ANN model. These improvements are due to the fact that the reaction rate constant shows non-linear correlations with the descriptors.

• Journal of the Korean Society of Manufacturing Process Engineers
• /
• v.18 no.2
• /
• pp.58-64
• /
• 2019

This paper describes the analysis of dynamic characteristics and prediction of the stiffness for the joint between structural members. In the process of deriving the governing equations, the stiffness values responsible for the moment and shear force were modelled by using linear and torsional springs in the middle of a clamped-clamped beam. The sensitivities of the natural frequency and modal assurance criterion were investigated as a function of the dimensionless linear and torsional spring stiffness. The reliability of the predictions for the linear and torsional stiffness values was verified by the inverse computations of the stiffness matrix. The predictive and exact theoretical stiffness values were compared for the stiffness element in the finite element formulation, and their results show an excellent correlation. It is strongly anticipated that although the proposed methodology is currently limited to the analytical utilization, it will provide a useful tool to estimate unknown joint stiffness values based on the experimental natural frequency and mode shape.

• Proceedings of the Korean Institute of Building Construction Conference
• /
• 2019.11a
• /
• pp.38-39
• /
• 2019

Environmental problems caused by GHG emitted by various industries are emerging around the world, and accordingly, relevant regulations are being applied by countries around the world. Korea is operating a carbon credit system that trades GHG in industry for money, which is expected to be applied to the construction industry. In addition, construction equipment using fossil fuels accounts for the largest portion of $CO_2$ emissions in the construction industry, and the importance of $CO_2$ reduction and prediction is increasing. However, there is a lack of data on the directly measured $CO_2$ emissions of construction equipment and there is no accurate methodology for measuring methods. Therefore, in this study, independent variables were derived based on the $CO_2$ emission data. In addition, multiple linear regression is performed for each independent variable to derive a predictive model of carbon dioxide emission by work type of construction equipment. It is expected that the construction process plan based on environmental factors in the construction industry can be established in the future.

• Asian Pacific Journal of Cancer Prevention
• /
• v.15 no.9
• /
• pp.4091-4094
• /
• 2014

Background: This study used the receiver operating characteristic curve (ROC) to analyze Surveillance, Epidemiology and End Results (SEER) bronchioaveolar carcinoma data to identify predictive models and potential disparity in outcomes. Materials and Methods: Socio-economic, staging and treatment factors were assessed. For the risk modeling, each factor was fitted by a Generalized Linear Model to predict cause specific survival. The area under the ROC was computed. Similar strata were combined to construct the most parsimonious models. A random sampling algorithm was used to estimate modeling errors. Risk of cause specific death was computed for the predictors for comparison. Results: There were 7,309 patients included in this study. The mean follow up time (S.D.) was 24.2 (20) months. Female patients outnumbered male ones 3:2. The mean (S.D.) age was 70.1 (10.6) years. Stage was the most predictive factor of outcome (ROC area of 0.76). After optimization, several strata were fused, with a comparable ROC area of 0.75. There was a 4% additional risk of death associated with lower county family income, African American race, rural residency and lower than 25% county college graduate. Radiotherapy had not been used in 2/3 of patients with stage III disease. Conclusions: There are socio-economic disparities in cause specific survival. Under-use of radiotherapy may have contributed to poor outcome. Improving education, access and rates of radiotherapy use may improve outcome.

• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1901-1904
• /
• 2003

MPC or model predictive control is representative of control methods which are able to handle physical constraints. Closed-loop stability can therefore be ensured only locally in the presence of constraints of this type. However, if the system is neutrally stable, and if the constraints are imposed only on the input, global aymptotic stability can be obtained; until recently, use of infinite horizons was thought to be inevitable in this case. A globally stabilizing finite-horizon MPC has lately been suggested for neutrally stable continuous-time systems using a non-quadratic terminal cost which consists of cubic as well as quadratic functions of the state. The idea originates from the so-called small gain control, where the global stability is proven using a non-quadratic Lyapunov function. The newly developed finite-horizon MPC employs the same form of Lyapunov function as the terminal cost, thereby leading to global asymptotic stability. A discrete-time version of this finite-horizon MPC is presented here. The proposed MPC algorithm is also coded using an SQP (Sequential Quadratic Programming) algorithm, and simulation results are given to show the effectiveness of the method.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2008.02a
• /
• pp.109-112
• /
• 2008

본 논문에서는 MPEG-4 HE-AAC의 SBR 부호기의 효율을 개선하기 위해 예측 SBR(Predictive-SBR)을 제안한다. SBR 부호기는 주부호기(core encoder)와 결합하여 적은 비트량으로 고주파 성분을 복원할 수 있게 한다. 본 논문에서는 SBR 데이터의 약 70%를 차지하는 포락선 정보를 부호화하는 방법을 개선하여 효율성을 높이고자 한다. 기존 SBR은 포락선 정보의 전송을 위해 다음과 같은 방법을 이용한다. 먼저 고주파 대역의 에너지를 스케일팩터 밴드 단위로 계산한다. 다음으로, 전송정보량의 감소를 위해 델타 코딩 방식을 이용하여 에너지 정보를 부호화한다. 본 논문에서는 SBR의 포락선 정보를 효과적으로 감축하기 위하여 고주파 대역의 에너지를 예측하는 방법을 이용한다. SBR 부호기의 입력 데이터가 SBR 복호기의 입력데이터와 동일하다는 가정 하에 선형 회귀(linear-regression) 기법을 이용하여 고주파 대역의 에너지를 추정한다. 그 후에 추정된 에너지와 원래의 고주파 대역 에너지의 오차를 델타 코딩을 이용하여 부호화한다. 정보를 전송할 때는 고주파 대역 에너지의 델타 코드와 예측 SBR에서 계산한 오차의 델타 코드 중 부호화에 필요한 비트량이 적은 방식을 선택하여 부호화하도록 한다. 그 결과 약 10% 정도의 정보량 감축 효과를 얻을 수 있다.