• Journal of Korea Water Resources Association
• /
• v.41 no.5
• /
• pp.527-544
• /
• 2008

Generalized Pareto distribution (GPD) is frequently applied in hydrologic extreme value analysis. The main objective of statistics of extremes is the prediction of rare events, and the primary problem has been the estimation of the threshold and the exceedances which were difficult without an accurate method of calculation. In this paper, to obtain the threshold or the exceedances, four methods were considered. For this comparison a GPD model was used to estimate parameters and quantiles for the seven durations (1, 2, 3, 6, 12, 18 and 24 hours) and the ten return periods (2, 3, 5, 10, 20, 30, 50, 70, 80 and 100 years). The parameters and quantiles of the three-parameter generalized Pareto distribution were estimated with three methods (MOM, ML and PWM). To estimate the degree of fit, three methods (K-S, CVM and A-D test) were performed and the relative root mean squared error (RRMSE) was calculated for a Monte Carlo generated sample. Then the performance of these methods were compared with the objective of identifying the best method from their number.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.23 no.1
• /
• pp.79-92
• /
• 2019

Titanium hydride potassium perchlorate (THPP) is one of the commonly utilized pyrotechnic materials in aerospace industries. The current study elucidates the effects of hygrothermal aging on the combustion of THPP experimentally. First, applying the Differential Scanning Calorimetry (DSC) and isocoversional method, both the delay of reaction start and decrease in maximum reaction rate were observed. The kinetics parameters tended to fluctuate depending the thermal reaction or intermediate product formation of THPP. Also, the oxidants decomposition and fuel oxidation phenomenon were discovered by X-ray photoelectron spectroscopy (XPS). The experimental heat from DSC data were verified as reasonable by comparing with the theoretical heat obtained utilizing both THPP formulation from XPS and NASA Chemical Equilibrium with Applications (CEA). Both data had identical variation trend, which expecially had the highest heat value at 10 weeks aged sample.

• Computers and Concrete
• /
• v.30 no.2
• /
• pp.99-111
• /
• 2022

In this study, deep learning and k-Nearest Neighbor (kNN) models were used to estimate the sorptivity and freeze-thaw resistance of self-compacting mortars (SCMs) having binary and ternary blends of mineral admixtures. Twenty-five environment-friendly SCMs were designed as binary and ternary blends of fly ash (FA) and silica fume (SF) except for control mixture with only Portland cement (PC). The capillary water absorption and freeze-thaw resistance tests were conducted for 91 days. It was found that the use of SF with FA as ternary blends reduced sorptivity coefficient values compared to the use of FA as binary blends while the presence of FA with SF improved freeze-thaw resistance of SCMs with ternary blends. The input variables used the models for the estimation of sorptivity were defined as PC content, SF content, FA content, sand content, HRWRA, water/cementitious materials (W/C) and freeze-thaw cycles. The input variables used the models for the estimation of sorptivity were selected as PC content, SF content, FA content, sand content, HRWRA, W/C and predefined intervals of the sample in water. The deep learning and k-NN models estimated the durability factor of SCM with 94.43% and 92.55% accuracy and the sorptivity of SCM was estimated with 97.87% and 86.14% accuracy, respectively. This study found that deep learning model estimated the sorptivity and durability factor of SCMs having binary and ternary blends of mineral admixtures higher accuracy than k-NN model.

• Exercise Science
• /
• v.23 no.1
• /
• pp.1-11
• /
• 2014

The purpose of this study was to develop and validate regression models to estimate maximal oxygen uptake (VO₂max) from the 20 m Progressive Shuttle Run Test (20 m PSRT) in Korean middle-school girls aged 13-15 years. The 20 m PSRT and VO₂max were assessed in a sample of 194 participants. The sample was randomly split into validation (n=127) and test-retest reliability (n=99, 32 out of 127 participants also performed validity test) groups. 127 participants performed a graded exercise test (GXT, stationary gas analyser) and the 20 m PSRT (portable gas analyser) once to develop a VO₂max prediction model and to analyze the validity of the modified 20 m PSRT protocol (starting at 7.5 km/h and increasing by 0.5 km/h every 1 min). 99 participants performed the 20 m PSRT twice for test-retest reliability purpose. Mean measured VO₂max (39.2±5.1 ml/kg/min) from the potable gas analyzer was significantly increased from that measured during the GXT from stationary gas analyzer (37.7±5.7 ml/kg/min, p=.001) using the modified 20 m PSRT protocol. But it was a narrow range (1.5 ml/kg/min). The measured VO₂max from the potable and stationary gas analyzers correlated at r=.88(p<.001). Test-retest of the 20 m PSRT yielded comparable results (Laps r=.88 & final speed r=.85). New regression equations were developed from present data to predict VO₂max for middle-school girls: y=.231×Laps-.311×weight(in kg)+46.201 (r=.74, SEE=4.29 ml/kg/min). It is concluded that (a) the modified 20 m PSRT protocol is a valid and reliable test and (b) this equation developed in this study provides valid estimates of VO₂max of Korean middle-school girl aged 13-15 years.

• Journal of Intelligence and Information Systems
• /
• v.26 no.1
• /
• pp.135-149
• /
• 2020

Artificial intelligences are changing world. Financial market is also not an exception. Robo-Advisor is actively being developed, making up the weakness of traditional asset allocation methods and replacing the parts that are difficult for the traditional methods. It makes automated investment decisions with artificial intelligence algorithms and is used with various asset allocation models such as mean-variance model, Black-Litterman model and risk parity model. Risk parity model is a typical risk-based asset allocation model which is focused on the volatility of assets. It avoids investment risk structurally. So it has stability in the management of large size fund and it has been widely used in financial field. XGBoost model is a parallel tree-boosting method. It is an optimized gradient boosting model designed to be highly efficient and flexible. It not only makes billions of examples in limited memory environments but is also very fast to learn compared to traditional boosting methods. It is frequently used in various fields of data analysis and has a lot of advantages. So in this study, we propose a new asset allocation model that combines risk parity model and XGBoost machine learning model. This model uses XGBoost to predict the risk of assets and applies the predictive risk to the process of covariance estimation. There are estimated errors between the estimation period and the actual investment period because the optimized asset allocation model estimates the proportion of investments based on historical data. these estimated errors adversely affect the optimized portfolio performance. This study aims to improve the stability and portfolio performance of the model by predicting the volatility of the next investment period and reducing estimated errors of optimized asset allocation model. As a result, it narrows the gap between theory and practice and proposes a more advanced asset allocation model. In this study, we used the Korean stock market price data for a total of 17 years from 2003 to 2019 for the empirical test of the suggested model. The data sets are specifically composed of energy, finance, IT, industrial, material, telecommunication, utility, consumer, health care and staple sectors. We accumulated the value of prediction using moving-window method by 1,000 in-sample and 20 out-of-sample, so we produced a total of 154 rebalancing back-testing results. We analyzed portfolio performance in terms of cumulative rate of return and got a lot of sample data because of long period results. Comparing with traditional risk parity model, this experiment recorded improvements in both cumulative yield and reduction of estimated errors. The total cumulative return is 45.748%, about 5% higher than that of risk parity model and also the estimated errors are reduced in 9 out of 10 industry sectors. The reduction of estimated errors increases stability of the model and makes it easy to apply in practical investment. The results of the experiment showed improvement of portfolio performance by reducing the estimated errors of the optimized asset allocation model. Many financial models and asset allocation models are limited in practical investment because of the most fundamental question of whether the past characteristics of assets will continue into the future in the changing financial market. However, this study not only takes advantage of traditional asset allocation models, but also supplements the limitations of traditional methods and increases stability by predicting the risks of assets with the latest algorithm. There are various studies on parametric estimation methods to reduce the estimated errors in the portfolio optimization. We also suggested a new method to reduce estimated errors in optimized asset allocation model using machine learning. So this study is meaningful in that it proposes an advanced artificial intelligence asset allocation model for the fast-developing financial markets.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.36 no.11
• /
• pp.1465-1471
• /
• 2007

Response surface methodology (RSM) was employed to optimize extraction conditions in order to find the maximal functional properties of fluid Cheonggukjang. Based on central composite design, a study plan was established with variations of microwave power, ethanol concentration, and extraction time. Regression analysis was applied to obtain a mathematical model. The maximum inhibitory of tyrosinase activity was found as 26.75% at the conditions of 30.56W microwave power, 2.40 g/mL of ratio of solvent to sample content and 10.00 min extraction time, respectively. The maximum superoxide dismutase (SOD)-like activity was 53.23% under the extraction conditions of 108.42 W, 4.38 g/mL and 7.84 min. Based on superimposition of three dimensional RSM with respect to extraction yield, inhibitory of tyrosinase activity and SOD-like activity obtained under the various extraction conditions, the optimum ranges of extraction conditions were found to be microwave power of $55{\sim}75$ W, ratio of solvent to sample content of $2{\sim}5$ g/mL and extraction time of $3.5{\sim}15$ min, respectively.

• Korean Journal of Food Science and Technology
• /
• v.26 no.5
• /
• pp.579-584
• /
• 1994

Near infrared reflectance(NIR) spectroscopy was used to determine the degree of retrogradation of cooked rice. Cooked rice samples were stored at $4^{\circ}C$ for 120 hours, and the degree of retrogradation was measured at every 6 hour during the storage time. Stored cooked rices were freeze-dried, milled and passed through a 100 mesh sieve. Enzymatic method using glucoamylase was used as reference method for the determination of the degree of retrogradation. Spectral differences due to retrogradation of cooked rice were observed at 1434, 1700, 1928, 2100, 2284 and 2320 nm. 32 samples of which moisture content were below 5% were used for calibration set, and 16 samples were used for validation set. High correlations were achieved between degree of retrogradation determined by conventional enzymatic method and by NIR with multiple correlation coefficient of 0.9753, and a standard error of calibration(SEC) of 3.64%. Comparable results were obtained with 3.91% of standard error of prediction(SEP), when the calibration equation was applied to independent group of samples of which moisture contents were in the range of calibration set. But when the calibration equation was applied to samples of which moisture contents were outer range of calibration set, SEP and bias were increased and correlation coefficient was decreased. The determination of degree of retrogradation was affected by sample moisture content. To determine degree of retrogradation of cooked rice by NIR using this calibration equation, it was suggested that sample moisture content should be controlled to below 5%.

• Bulletin of the Society of Naval Architects of Korea
• /
• v.27 no.2
• /
• pp.1-12
• /
• 1990

Interactions between a propeller and vortex system contained in a ship stern flow is treated theoretically. A new formulation to determine the effective velocity distributions is developed, which may be immediately applicable to the design and analysis of compound propulsors under the influence of severe vortical cross-flows around ship stern. An axisymmetric shear flow is represented by a system of ring vortices and the axial variation of the stream lines due to the action of propeller is represented by a cubic function. The strengths of ring vortices, which are varying along the stream lines, are determined by the conservation of angular momentum. Two simplified effective velocity models are proposed to confirm the theory. Sample calculations using the simplified models are made to compare with the results by other investigators.

• Tunnel and Underground Space
• /
• v.17 no.2 s.67
• /
• pp.119-127
• /
• 2007

Based on the critical plane approach, a methodology far predicting the anisotropic strength ot transversely isotropic rock is Proposed. It is assumed that the rock failure is governed by Hoek-Brown failure criterion. In order to establish an anisotropic failure function, Mohr envelope equivalent to the original Hoek-Brown criterion is used and the strength parameters m, s are expressed as scalar functions of orientation. The conjugate gradient method, which is one of the robust optimization techniques, is applied to the failure function for searching the orientation giving the maximum value of the anisotropic function. While most of the existing anisotropic strength models can be applied only when the stress condition is the same as that of conventional triaxial compression test, the proposed model can be applied to the general 3-dimensional stress conditions. Through the simulation of triaxial compression tests for transversely isotropic rock sample, the validity of the proposed method is investigated by comparing the predicted triaxial strengths and inclinations of failure plane.

• Journal of the Korean Society of School Health
• /
• v.36 no.3
• /
• pp.113-125
• /
• 2023

Purpose: The purpose of this study is to develop a model for predicting mental health risk among adolescents based on health behavior information by employing a hybrid machine learning method. Methods: The study analyzed data of 51,850 domestic middle and high school students from 2022 Youth Health Behavior Survey conducted by the Korea Disease Control and Prevention Agency. Firstly, mental health risk levels (stress perception, suicidal thoughts, suicide attempts, suicide plans, experiences of sadness and despair, loneliness, and generalized anxiety disorder) were classified using the k-mean unsupervised learning technique. Secondly, demographic factors (family economic status, gender, age), academic performance, physical health (body mass index, moderate-intensity exercise, subjective health perception, oral health perception), daily life habits (sleep time, wake-up time, smartphone use time, difficulty recovering from fatigue), eating habits (consumption of high-caffeine drinks, sweet drinks, late-night snacks), violence victimization, and deviance (drinking, smoking experience) data were input to develop a random forest model predicting mental health risk, using logistic and XGBoosting. The model and its prediction performance were compared. Results: First, the subjects were classified into two mental health groups using k-mean unsupervised learning, with the high mental health risk group constituting 26.45% of the total sample (13,712 adolescents). This mental health risk group included most of the adolescents who had made suicide plans (95.1%) or attempted suicide (96.7%). Second, the predictive performance of the random forest model for classifying mental health risk groups significantly outperformed that of the reference model (AUC=.94). Predictors of high importance were 'difficulty recovering from daytime fatigue' and 'subjective health perception'. Conclusion: Based on an understanding of adolescent health behavior information, it is possible to predict the mental health risk levels of adolescents and make interventions in advance.