• Journal of The Korean Astronomical Society
• /
• v.52 no.6
• /
• pp.217-225
• /
• 2019

We apply a modified Convolutional Neural Network (CNN) model in conjunction with transfer learning to predict whether an active region (AR) would produce a ≥C-class or ≥M-class flare within the next 24 hours. We collect line-of-sight magnetogram samples of ARs provided by the SHARP from May 2010 to September 2018, which is a new data product from the HMI onboard the SDO. Based on these AR samples, we adopt the approach of shuffle-and-split cross-validation (CV) to build a database that includes 10 separate data sets. Each of the 10 data sets is segregated by NOAA AR number into a training and a testing data set. After training, validating, and testing our model, we compare the results with previous studies using predictive performance metrics, with a focus on the true skill statistic (TSS). The main results from this study are summarized as follows. First, to the best of our knowledge, this is the first time that the CNN model with transfer learning is used in solar physics to make binary class predictions for both ≥C-class and ≥M-class flares, without manually engineered features extracted from the observational data. Second, our model achieves relatively high scores of TSS = 0.640±0.075 and TSS = 0.526±0.052 for ≥M-class prediction and ≥C-class prediction, respectively, which is comparable to that of previous models. Third, our model also obtains quite good scores in five other metrics for both ≥C-class and ≥M-class flare prediction. Our results demonstrate that our modified CNN model with transfer learning is an effective method for flare forecasting with reasonable prediction performance.

• Journal of the Korean Institute of Gas
• /
• v.16 no.1
• /
• pp.26-32
• /
• 2012

The autoignition temperature is one of the most important physical properties used to determine the flammability characteristics of chemical substances. Despite the needs of the experimental autoignition temperature data for the design of chemical plants, it is not easy to get the data. This study have built and compared partial least squares (PLS) and support vector machine (SVM) models to predict the autoignition temperatures of 503 organic compounds out of DIPPR 801. As the independent variables of the models, 59 functional groups were chosen based on the group contribution method. The prediction errors calculated from cross-validation were employed to determine the optimal parameters of two models. And, particle swarm optimization was used to get three parameters of SVM model. The PLS and SVM results of the average absolute errors for the whole data range from 58.59K and 29.11K, respectively, indicating that the predictive ability of the SVM is much superior than PLS.

• Journal of Nutrition and Health
• /
• v.33 no.8
• /
• pp.864-872
• /
• 2000

This study intended to evaluate the validity of the simple nutrition screening test that had been developed with the elderly living in Cheongju as a subject. Nutrition screening score(NSS) and reference standards for nutritional and health status(nutrient intakes, mean adequacy ratio, perceive health, and serum albumin, hematocrit, and hemoglobin) were estimated by using the date obtained in 1996 from the 174 elderly living in Taejon, Statistical analysis showed significant correlations between mean adequacy ratio(MAR) and NSS(r=0.341) and also between NSS and biological indices such as albumin and hematocrit, Around 65-75% of the elderly with perceive health and low level of serum albumin, hemoglobin and hematocrit had NSS$\leq$ll. Sensitivity, specificity, and positive predictive values(PPV) were calculated from the crosstabulation of the three categories of NSS(high, moderate, and low nutritional risk) and low categories MAR(< 0.75, undernutrition;$\geq$0.75, normal) to validate the cut-off point for high or low nutritional risk by NSS. It was suggested that point l1 was appropriate as a criterion to determine high risk of undernutrition, but point 16 was better than 17 as criterion to determine low nutritional risk in the Taejon elderly. When point ll was used as a criterion of high nutritional risk, sensitivity, specificity, and PPV are 59.5, 60.5 and 82.1 respectively. When point 16 was used as a criterion of low nutritional risk, sensitivity, specificity, and PPV are 25.6, 95.4, and 64.7%, respectively. In conclusion, nutrition screening test that had been developed can be a simple, easy, and proper instrument to classify the high risk group of undernutrition. A further validation study seems to be required among other groups of individuals for the screening test to the finalized as a more valid instrument identifying Korean elderly at nutrition and health risk(Korean J Nutrition 33(8) : 864-872, 2000)

• Journal of The Korean Society of Agricultural Engineers
• /
• v.59 no.5
• /
• pp.93-99
• /
• 2017

A model was developed using Artificial Neural Networks (ANNs) based on Principal Component Analysis (PCA), to accurately predict the air humidity inside an experimental greenhouse located in Daegu (latitude $35.53^{\circ}N$, longitude $128.36^{\circ}E$, and altitude 48 m), South Korea. The weather parameters, air temperature, relative humidity, solar radiation, and carbon dioxide inside and outside the greenhouse were monitored and measured by mounted sensors. Through the PCA of the data samples, three main components were used as the input data, and the measured inside humidity was used as the output data for the ALYUDA forecaster software of the ANN model. The Nash-Sutcliff Model Efficiency Coefficient (NSE) was used to analyze the difference between the experimental and the simulated results, in order to determine the predictive power of the ANN software. The results obtained revealed the variables that affect the inside air humidity through a sensitivity analysis graph. The measured humidity agreed well with the predicted humidity, which signifies that the model has a very high accuracy and can be used for predictions based on the computed $R^2$ and NSE values for the training and validation samples.

• Toxicological Research
• /
• v.32 no.1
• /
• pp.47-56
• /
• 2016

The identification of biomarkers for the early detection of acute kidney injury (AKI) is clinically important. Acute kidney injury (AKI) in critically ill patients is closely associated with increased morbidity and mortality. Conventional biomarkers, such as serum creatinine (SCr) and blood urea nitrogen (BUN), are frequently used to diagnose AKI. However, these biomarkers increase only after significant structural damage has occurred. Recent efforts have focused on identification and validation of new noninvasive biomarkers for the early detection of AKI, prior to extensive structural damage. Furthermore, AKI biomarkers can provide valuable insight into the molecular mechanisms of this complex and heterogeneous disease. Our previous study suggested that pyruvate kinase M2 (PKM2), which is excreted in the urine, is a sensitive biomarker for nephrotoxicity. To appropriately and optimally utilize PKM2 as a biomarker for AKI requires its complete characterization. This review highlights the major studies that have addressed the diagnostic and prognostic predictive power of biomarkers for AKI and assesses the potential usage of PKM2 as an early biomarker for AKI. We summarize the current state of knowledge regarding the role of biomarkers and the molecular and cellular mechanisms of AKI. This review will elucidate the biological basis of specific biomarkers that will contribute to improving the early detection and diagnosis of AKI.

• The korean journal of orthodontics
• /
• v.46 no.5
• /
• pp.331-342
• /
• 2016

Objective: To evaluate the accuracy and reliability of the diagnostic tools available for assessing maxillary transverse deficiencies. Methods: An electronic search of three databases was performed from their date of establishment to April 2015, with manual searching of reference lists of relevant articles. Articles were considered for inclusion if they reported the accuracy or reliability of a diagnostic method or evaluation technique for maxillary transverse dimensions in mixed or permanent dentitions. Risk of bias was assessed in the included articles, using the Quality Assessment of Diagnostic Accuracy Studies tool-2. Results: Nine articles were selected. The studies were heterogeneous, with moderate to low methodological quality, and all had a high risk of bias. Four suggested that the use of arch width prediction indices with dental cast measurements is unreliable for use in diagnosis. Frontal cephalograms derived from cone-beam computed tomography (CBCT) images were reportedly more reliable for assessing intermaxillary transverse discrepancies than posteroanterior cephalograms. Two studies proposed new three-dimensional transverse analyses with CBCT images that were reportedly reliable, but have not been validated for clinical sensitivity or specificity. No studies reported sensitivity, specificity, positive or negative predictive values or likelihood ratios, or ROC curves of the methods for the diagnosis of transverse deficiencies. Conclusions: Current evidence does not enable solid conclusions to be drawn, owing to a lack of reliable high quality diagnostic studies evaluating maxillary transverse deficiencies. CBCT images are reportedly more reliable for diagnosis, but further validation is required to confirm CBCT's accuracy and diagnostic superiority.

• Interdisciplinary Bio Central
• /
• v.3 no.4
• /
• pp.15.1-15.11
• /
• 2011

Background: Histone deacetylase (HDAC) 8 is one of its family members catalyzes the removal of acetyl groups from N-terminal lysine residues of histone proteins thereby restricts transcription factors from being expressed. Inhibition of HDAC8 has become an emerging and effective anti-cancer therapy for various cancers. Application computational methodologies may result in identifying the key components that can be used in developing future potent HDAC8 inhibitors. Results: Facilitating the discovery of novel and potential chemical scaffolds as starting points in the future HDAC8 inhibitor design, quantitative structure-activity relationship models were generated with 30 training set compounds using genetic function approximation (GFA) and Bayesian algorithms. Six GFA models were selected based on the significant statistical parameters calculated during model development. A Bayesian model using fingerprints was developed with a receiver operating characteristic curve cross-validation value of 0.902. An external test set of 54 diverse compounds was used in validating the models. Conclusions: Finally two out of six models based on their predictive ability over the test set compounds were selected as final GFA models. The Bayesian model has displayed a high classifying ability with the same test set compounds and the positively and negatively contributing molecular fingerprints were also unveiled by the model. The effectively contributing physicochemical properties and molecular fingerprints from a set of known HDAC8 inhibitors were identified and can be used in designing future HDAC8 inhibitors.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.60 no.2
• /
• pp.398-406
• /
• 2011

In this paper, we introduce a new architecture of PSO-based Polynomial Neural Networks (PNN) and discuss its comprehensive design methodology. The conventional PNN is based on a extended Group Method of Data Handling (GMDH) method, and utilized the polynomial order (viz. linear, quadratic, and modified quadratic) as well as the number of node inputs fixed (selected in advance by designer) at Polynomial Neurons located in each layer through a growth process of the network. Moreover it does not guarantee that the conventional PNN generated through learning results in the optimal network architecture. The PSO-based PNN results in a structurally optimized structure and comes with a higher level of flexibility that the one encountered in the conventional PNN. The PSO-based design procedure being applied at each layer of PNN leads to the selection of preferred PNs with specific local characteristics (such as the number of input variables, input variables, and the order of the polynomial) available within the PNN. In the sequel, two general optimization mechanisms of the PSO-based PNN are explored: the structural optimization is realized via PSO whereas in case of the parametric optimization we proceed with a standard least square method-based learning. To evaluate the performance of the PSO-based PNN, the model is experimented with using Gas furnace process data, and pH neutralization process data. For the characteristic analysis of the given entire data with non-linearity and the construction of efficient model, the given entire system data is partitioned into two type such as Division I(Training dataset and Testing dataset) and Division II(Training dataset, Validation dataset, and Testing dataset). A comparative analysis shows that the proposed PSO-based PNN is model with higher accuracy as well as more superb predictive capability than other intelligent models presented previously.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• v.1
• /
• pp.395-400
• /
• 2006

Among the various error sources in positioning and navigation, the paper focuses on the modeling and prediction of receiver clock bias and then tries to achieve positioning based on simulated and predicted clock bias. With the SA off, it is possible to model receiver clock bias more accurately. We selected several types of GNSS receivers for test using ARMA model. To facilitate prediction with short and limited sample pseudorange observations, AR and ARMA are compared, and the improved AR model is presented to model and predict receiver clock bias based on previous solutions. Our work extends to clock bias prediction and positioning based on predicted clock bias using only 3 satellites that is usually the case under urban canyon situation. In contrast to previous experiences, we find that a receiver clock bias can be well modeled using adopted ARMA model. Test has been done on various types of GNSS receivers to show the validation of developed model. To further develop this work, we compare solution conditions in terms of DOP values when point positioning is conducted using 3 satellites to simulate urban positioning environment. When condition allows, height component is derived from other ways and can be set as known values. Given this condition, location is possible using less than 2 GNSS satellites with fixed height. Solution condition is also discussed for this background using mode of constrained positioning. We finally suggest an effective predictive time span based on our test exploration under varied conditions.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.1
• /
• pp.13-21
• /
• 2020

Surrogate models have been used for the rapid estimation of six-DOF aerodynamic coefficients in the context of the design and control of a missile. For this end, we may generate highly accurate surrogate models with a multitude of aerodynamic data obtained from wind tunnel tests (WTTs); however, this approach is time-consuming and expensive. Thus, we aim to swiftly predict aerodynamic coefficients via co-Kriging using a few WTT data along with plenty of computational fluid dynamics (CFD) data. To demonstrate the excellence of co-Kriging models based on both WTT and CFD data, we first generated two surrogate models: co-Kriging models with CFD data and Kriging models without the CFD data. Afterwards, we carried out numerical validation and examined predictive trends to compare the two different surrogate models. As a result, we found that the co-Kriging models produced more accurate aerodynamic coefficients than the Kriging models thanks to the assistance of CFD data.