• Journal of The Korean Society of Emergency Medicine
• /
• v.29 no.5
• /
• pp.455-464
• /
• 2018

Objective: Machine learning is not yet widely used in the medical field. Therefore, this study was conducted to compare the performance of preexisting severity prediction models and machine learning based models (random forest [RF], gradient boosting [GB]) for mortality prediction in pneumonia patients. Methods: We retrospectively collected data from patients who visited the emergency department of a tertiary training hospital in Seoul, Korea from January to March of 2015. The Pneumonia Severity Index (PSI) and Sequential Organ Failure Assessment (SOFA) scores were calculated for both groups and the area under the curve (AUC) for mortality prediction was computed. For the RF and GB models, data were divided into a test set and a validation set by the random split method. The training set was learned in RF and GB models and the AUC was obtained from the validation set. The mean AUC was compared with the other two AUCs. Results: Of the 536 investigated patients, 395 were enrolled and 41 of them died. The AUC values of PSI and SOFA scores were 0.799 (0.737-0.862) and 0.865 (0.811-0.918), respectively. The mean AUC values obtained by the RF and GB models were 0.928 (0.899-0.957) and 0.919 (0.886-0.952), respectively. There were significant differences between preexisting severity prediction models and machine learning based models (P<0.001). Conclusion: Classification through machine learning may help predict the mortality of pneumonia patients visiting the emergency department.

• 지반과기술
• /
• v.3 no.3
• /
• pp.15-22
• /
• 2006

The optimum blasting pattern to excavate a quarry efficiently and economically can be determined based on the minimum production cost, which is generally estimated according to rock fragmentation. Therefore, it is a critical problem to predict fragment size distribution of blasted rocks over an entire quarry. By comparing various prediction models, it can be ascertained that the result obtained from Kuz-Ram model relatively coincides with that of field measurements. Kuz-Ram model uses the concept of rock factor to signify conditions of rock mass such as block size, rock jointing, strength and others. For the evaluation of total production cost, it is imperative to estimate 3-D spatial distribution of rock factor for the entire quarry. In this study, a sequential indicator simulation technique is adopted for estimation of spatial distribution of rock factor due to its higher reproducibility of spatial variability and distribution models than Kriging methods. Further, this can reduce the uncertainty of predictor using distribution information of sample data. The entire quarry is classified into three types of rock mass and optimum blasting pattern is proposed for each type based on 3-D spatial distribution of rock factor. In addition, plane maps of rock factor distribution for each ground level are provided to estimate production costs for each process and to make a plan for an optimum blasting pattern.

• Explosives and Blasting
• /
• v.24 no.1
• /
• pp.1-8
• /
• 2006

Recently, most of limestone quarries have been not mined by open-pit mining but by underground excavation to reduce environmental pollution. As a result, the size of underground galleries became bigger to maintain mass-production close to open-pit mining. However, the scale of pillars and galleries as well as the excavation methods may induce a few adverse problems for the stability of a mined gallery. In this study, the nomogram analysis and the prediction of rock damage zone induced by blasting were carried out. The testing conditions include concurrent blasting of two adjacent galleries, concurrent blasting of a transport drift and a inclined shaft, sequential blasting of two galleries, and separate blasting for each gallery. For each testing condition, blast vibration velocity was measured and analyzed. From the prediction formulas for blast vibration velocity derived in this study, the maximum depth of rock damage zone induced by blasting were also predicted.

• Transactions of Materials Processing
• /
• v.24 no.6
• /
• pp.395-399
• /
• 2015

A microstructure-based finite element simulation was conducted to predict the plastic strain ratio (R-value) of a dual-phase (DP) steel. The representative volume elements (RVEs) concept was adopted for the image-based FE modeling and a 3D model was constructed using sequential 2D images. Each phase was considered with the von-Mises yield criterion and the Swift model. The Swift parameters were defined by the empirical equations based on the chemical composition. The developed model was applied to analyze the effect of residual stress on the R-value and stress distribution. In order to consider the residual stress development after cold rolling, 10 % compression was applied in the thickness direction and unloaded before the tensile stress was applied in the rolling direction. The results showed a reasonable prediction for the R-value evolution: a sharp increase at small strains was well described and a transition followed in the downward direction. The R-value evolution was analyzed using the stress distribution change on the π-plane

• Journal of the Korean Society of Safety
• /
• v.7 no.2
• /
• pp.71-79
• /
• 1992

This paper presents a least-square algorithms of lattice structures and their use for adaptive prediction of time series generated from the dynamic system. As the view point of adaptive prediction, a new method of Identification of dynamic characteristics by means of estimating the parameters of linear auto regressive model is proposed. The fast convergence of adaptive lattice algorithms is seen to be due to the orthogonalization and decoupling properties of the lattice. The superiority of the least-square lattice is verified by computer simulation, then predictor coefficients are computed from the linear sequential time data. For the application to the dynamic characteristic analysis of unknown system, the transfer function of ideal system represented in frquency domain and the estimated one obtained by predicted coefficients are compared. Using the proposed method, the damping ratio and the natural frequency of a dynamic structure subjected to random excitations can be estimated. It is expected that this method will be widely applicable to other technical dynamic problem in which estimation of damping ratio and fundamental vibration modes are required.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.22 no.2
• /
• pp.278-286
• /
• 2019

Conventional methods for selecting jamming techniques in electronic warfare are based on libraries in which a list of jamming techniques for radar signals is recorded. However, the choice of jamming techniques by the library is limited when modified signals are received. In this paper, we propose a method to predict the jamming technique for radar signals by using deep learning methods. Long short-term memory(LSTM) is a deep running method which is effective for learning the time dependent relationship in sequential data. In order to determine the optimal LSTM model structure for jamming technique prediction, we test the learning parameter values that should be selected, such as the number of LSTM layers, the number of fully-connected layers, optimization methods, the size of the mini batch, and dropout ratio. Experimental results demonstrate the competent performance of the LSTM model in predicting the jamming technique for radar signals.

• Journal of Wellbeing Management and Applied Psychology
• /
• v.6 no.2
• /
• pp.27-31
• /
• 2023

Purpose: Establishment of a real-time monitoring system for odor control in traditional markets in Gangwon-do and a system for linking prevention facilities. Research design, data and methodology: Build server and system logic based on data through real-time monitoring device (sensor-based). A temporary data generation program for deep learning is developed to develop a model for odor data. Results: A REST API was developed for using the model prediction service, and a test was performed to find an algorithm with high prediction probability and parameter values optimized for learning. In the deep learning algorithm for AI modeling development, Pandas was used for data analysis and processing, and TensorFlow V2 (keras) was used as the deep learning library. The activation function was swish, the performance of the model was optimized for Adam, the performance was measured with MSE, the model method was Functional API, and the model storage format was Sequential API (LSTM)/HDF5. Conclusions: The developed system has the potential to effectively monitor and manage odors in traditional markets. By utilizing real-time data, the system can provide timely alerts and facilitate preventive measures to control and mitigate odors. The AI modeling component enhances the system's predictive capabilities, allowing for proactive odor management.

• Geomechanics and Engineering
• /
• v.34 no.1
• /
• pp.29-41
• /
• 2023

The 3D geospatial modeling of geotechnical information can aid in understanding the geotechnical characteristic values of the continuous subsurface at construction sites. In this study, a geostatistical optimization model for the three-dimensional (3D) mapping of subsurface stratification and the SPT-N value based on a trial-and-error rule was developed and applied to a dam emergency spillway site in South Korea. Geospatial database development for a geotechnical investigation, reconstitution of the target grid volume, and detection of outliers in the borehole dataset were implemented prior to the 3D modeling. For the site-specific subsurface stratification of the engineering geo-layer, we developed an integration method for the borehole and geophysical survey datasets based on the geostatistical optimization procedure of ordinary kriging and sequential Gaussian simulation (SGS) by comparing their cross-validation-based prediction residuals. We also developed an optimization technique based on SGS for estimating the 3D geometry of the SPT-N value. This method involves quantitatively testing the reliability of SGS and selecting the realizations with a high estimation accuracy. Boring tests were performed for validation, and the proposed method yielded more accurate prediction results and reproduced the spatial distribution of geotechnical information more effectively than the conventional geostatistical approach.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.2
• /
• pp.77-86
• /
• 2020

In this study, the Eulerian/Lagrangian one-way coupling method is proposed to predict flow noise due to Blade-Tip Vortex Cavitation (BTVC). The proposed method consists of four sequential steps: flow field simulation using Computational Fluid Dynamics (CFD) techniques, reconstruction of wing-tip vortex using vortex model, generation of BTVC using bubble dynamics model and acoustic wave prediction using the acoustic analogy. Because the CFD prediction of tip vortex structure generally suffers from severe under-prediction of its strength along the steamwise direction due to the intrinsic numerical damping of CFD schemes and excessive turbulence intensity, the wing-tip vortex along the freestream direction is regenerated by using the vortex modeling. Then, the bubble dynamics model based on the Rayleigh-Plesset equation was employed to simulate the generation and variation of BTVC. Finally, the flow noise due to BTVC is predicted by modeling each of spherical bubbles as a monople source whose strength is proportional to the rate of time-variation of bubble volume. The validity of the proposed numerical methods is confirmed by comparing the predicted results with the measured data.

• Journal of Digital Convergence
• /
• v.14 no.11
• /
• pp.267-273
• /
• 2016

Creation and Sharing of information which is structured data as well as various unstructured data. makes progress actively through the spread of mobile. Recently, Big Data extracts the semantic information from SNS and data mining is one of the big data technique. Especially, the general emotion analysis that expresses the collective intelligence of the masses is utilized using large and a variety of materials. In this paper, we propose the emotion prediction system architecture which extracts the significant keywords from social network paragraphs using n-gram and Korean morphological analyzer, and predicts the emotion using SVM and these extracted emotion features. The proposed system showed 82.25% more improved recall rate in average than previous systems and it will help extract the semantic keyword using morphological analysis.