• Journal of Korea Water Resources Association
• /
• v.55 no.6
• /
• pp.405-419
• /
• 2022

Flood frequency analysis commonly used to design the hydraulic structures to minimize flood damage includes uncertainty. Therefore, the most appropriate design flood within a uncertainty should be selected in the final stage of a hydraulic structure, but related studies were rarely carried out. The total expected cost function introduced into the flood frequency analysis is a new approach for determining the optimal design flood. This procedure has been used as UNCODE (UNcertainty COmpliant DEsign), but the application has not yet been introduced in South Korea. This study introduced the mathematical procedure of UNCODE and calculated the optimal design flood using the annual maximum inflow of hydroelectric dams located in the Bukhan River system and results were compared with that of the existing flood frequency. The parameter uncertainty was considered in the total expected cost function using the Gumbel and the GEV distribution, and the Metropolis-Hastings algorithm was used to sample the parameters. In this study, cost function and damage function were assumed to be a first-order linear function. It was found that the medians of the optimal design flood for 4 Hydroelectric dams, 2 probability distributions, and 2 return periods were calculated to be somewhat larger than the design flood by the existing flood frequency analysis. In the future, it is needed to develop the practical approximated procedure to UNCODE.

• Journal of Intelligence and Information Systems
• /
• v.28 no.4
• /
• pp.309-328
• /
• 2022

This study empirically analyzed a Korean pre-trained language models (PLMs) designed for natural language generation. The performance of two PLMs - BART and GPT - at the task of abstractive text summarization was compared. To investigate how performance depends on the characteristics of the inference data, ten different document types, containing six types of informational content and creation content, were considered. It was found that BART (which can both generate and understand natural language) performed better than GPT (which can only generate). Upon more detailed examination of the effect of inference data characteristics, the performance of GPT was found to be proportional to the length of the input text. However, even for the longest documents (with optimal GPT performance), BART still out-performed GPT, suggesting that the greatest influence on downstream performance is not the size of the training data or PLMs parameters but the structural suitability of the PLMs for the applied downstream task. The performance of different PLMs was also compared through analyzing parts of speech (POS) shares. BART's performance was inversely related to the proportion of prefixes, adjectives, adverbs and verbs but positively related to that of nouns. This result emphasizes the importance of taking the inference data's characteristics into account when fine-tuning a PLMs for its intended downstream task.

• Journal of Intelligence and Information Systems
• /
• v.29 no.1
• /
• pp.107-119
• /
• 2023

The technology that recognizes a soldier's motion and movement status has recently attracted large attention as a combination of wearable technology and artificial intelligence, which is expected to upend the paradigm of troop management. The accuracy of state determination should be maintained at a high-end level to make sure of the expected vital functions both in a training situation; an evaluation and solution provision for each individual's motion, and in a combat situation; overall enhancement in managing troops. However, when input data is given as a timer series or sequence, existing feedforward networks would show overt limitations in maximizing classification performance. Since human behavior data (3-axis accelerations and 3-axis angular velocities) handled for military motion recognition requires the process of analyzing its time-dependent characteristics, this study proposes a high-performance data-driven classifier which utilizes the long-short term memory to identify the order dependence of acquired data, learning to classify eight representative military operations (Sitting, Standing, Walking, Running, Ascending, Descending, Low Crawl, and High Crawl). Since the accuracy is highly dependent on a network's learning conditions and variables, manual adjustment may neither be cost-effective nor guarantee optimal results during learning. Therefore, in this study, we optimized hyperparameters using Bayesian optimization for maximized generalization performance. As a result, the final architecture could reduce the error rate by 62.56% compared to the existing network with a similar number of learnable parameters, with the final accuracy of 98.39% for various military operations.

• Journal of Internet of Things and Convergence
• /
• v.9 no.3
• /
• pp.93-98
• /
• 2023

In this paper, an artificial neural network based on simulated annealing was constructed. The mapping relationship between the parameters of micro-scaled flow channels electrochemical machining and the channel shape was established by training the samples. The depth and width of micro-scaled flow channels electrochemical machining on stainless steel surface were predicted, and the flow channels experiment was carried out with pulse power supply in NaNO3 solution to verify the established network model. The results show that the depth and width of the channel predicted by the simulated annealing artificial neural network with "4-7-2" structure are very close to the experimental values, and the error is less than 5.3%. The predicted and experimental data show that the etching degree in the process of channels electrochemical machining is closely related to voltage and current density. When the voltage is less than 5V, a "small island" is formed in the channel; When the voltage is greater than 40V, the lateral etching of the channel is relatively large, and the "dam" between the channels disappears. When the voltage is 25V, the machining morphology of the channel is the best.

• Advances in nano research
• /
• v.16 no.4
• /
• pp.375-394
• /
• 2024

The extensive utilization of concrete has given rise to environmental concerns, specifically concerning the depletion of river sand. To address this issue, waste deposits can provide manufactured-sand (MS) as a substitute for river sand. The objective of this study is to explore the application of machine learning techniques to facilitate the production of manufactured-sand concrete (MSC) containing stone nano-powder through estimating the splitting tensile strength (STS) containing compressive strength of cement (CSC), tensile strength of cement (TSC), curing age (CA), maximum size of the crushed stone (Dmax), stone nano-powder content (SNC), fineness modulus of sand (FMS), water to cement ratio (W/C), sand ratio (SR), and slump (S). To achieve this goal, a total of 310 data points, encompassing nine influential factors affecting the mechanical properties of MSC, are collected through laboratory tests. Subsequently, the gathered dataset is divided into two subsets, one for training and the other for testing; comprising 90% (280 samples) and 10% (30 samples) of the total data, respectively. By employing the generated dataset, novel models were developed for evaluating the STS of MSC in relation to the nine input features. The analysis results revealed significant correlations between the CSC and the curing age CA with STS. Moreover, when delving into sensitivity analysis using an empirical model, it becomes apparent that parameters such as the FMS and the W/C exert minimal influence on the STS. We employed various loss functions to gauge the effectiveness and precision of our methodologies. Impressively, the outcomes of our devised models exhibited commendable accuracy and reliability, with all models displaying an R-squared value surpassing 0.75 and loss function values approaching insignificance. To further refine the estimation of STS for engineering endeavors, we also developed a user-friendly graphical interface for our machine learning models. These proposed models present a practical alternative to laborious, expensive, and complex laboratory techniques, thereby simplifying the production of mortar specimens.

• Korean Journal of Radiology
• /
• v.25 no.2
• /
• pp.189-198
• /
• 2024

Objective: To investigate the prognostic utility of radiomics features extracted from ¹⁸F-fluorodeoxyglucose (FDG) PET/CT combined with clinical factors and metabolic parameters in predicting progression-free survival (PFS) and overall survival (OS) in individuals diagnosed with extranodal nasal-type NK/T cell lymphoma (ENKTCL). Materials and Methods: A total of 126 adults with ENKTCL who underwent ¹⁸F-FDG PET/CT examination before treatment were retrospectively included and randomly divided into training (n = 88) and validation cohorts (n = 38) at a ratio of 7:3. Least absolute shrinkage and selection operation Cox regression analysis was used to select the best radiomics features and calculate each patient's radiomics scores (RadPFS and RadOS). Kaplan-Meier curve and Log-rank test were used to compare survival between patient groups risk-stratified by the radiomics scores. Various models to predict PFS and OS were constructed, including clinical, metabolic, clinical + metabolic, and clinical + metabolic + radiomics models. The discriminative ability of each model was evaluated using Harrell's C index. The performance of each model in predicting PFS and OS for 1-, 3-, and 5-years was evaluated using the time-dependent receiver operating characteristic (ROC) curve. Results: Kaplan-Meier curve analysis demonstrated that the radiomics scores effectively identified high- and low-risk patients (all P < 0.05). Multivariable Cox analysis showed that the Ann Arbor stage, maximum standardized uptake value (SUVmax), and RadPFS were independent risk factors associated with PFS. Further, β2-microglobulin, Eastern Cooperative Oncology Group performance status score, SUVmax, and RadOS were independent risk factors for OS. The clinical + metabolic + radiomics model exhibited the greatest discriminative ability for both PFS (Harrell's C-index: 0.805 in the validation cohort) and OS (Harrell's C-index: 0.833 in the validation cohort). The time-dependent ROC analysis indicated that the clinical + metabolic + radiomics model had the best predictive performance. Conclusion: The PET/CT-based clinical + metabolic + radiomics model can enhance prognostication among patients with ENKTCL and may be a non-invasive and efficient risk stratification tool for clinical practice.

• Journal of Korean Society of Occupational and Environmental Hygiene
• /
• v.34 no.1
• /
• pp.77-84
• /
• 2024

Objectives: Interventions promoting physical exercise and healthy habits in workplaces have been shown to be effective in reducing risk factors for metabolic syndrome. This study was conducted to examine the effects of an individualized conditioning exercise program of IT company office workers with or at higher risk of metabolic syndrome. Methods: A total of 444 IT company office workers with or at higher risk of metabolic syndrome participated in a 3-month conditioning exercise program. Body composition data using bioelectrical impedance analysis and cardiopulmonary data using cardiopulmonary exercise testing from 53 individuals (mean age: 34.8 ± 7.1 years, sex : 21% female, height : 170.4 ± 6.8 cm, weight : 75.2±12.2 kg, body mass index : 25.8±3.3 kg/m²) who have successfully completed pre-test, intervention, and post-test were analyzed. The 12 weeks intervention encompassed: (1) health counseling (2) supervised exercise(endurance-based, aerobic exercise, or circuit training once a week for 50 minutes at heart rate reserve(HRR) of 77-95%) (3) self-directed exercise and biweekly health screening checks. Results: The results indicated a significant decrease in body weight, body fat mass and body mass index, respectively. Moreover, VO₂peak, AT VO₂ and AT Time significantly improved, respectively. Resting blood pressure(SBP/DBP) showed positive changes but were not statistically significant. We observed the correlation between characteristics of participants and rate of changes in cardiopulmonary outcomes of participants, there are no significant correlation. These results indicate positive changes in body composition and cardiorespiratory fitness parameters following individualized conditioning exercise program. Conclusions: Individualized workplace exercise program for preventing metabolic syndrome can lead to improvements in body composition and cardiorespiratory fitness.

• Journal of Korea Water Resources Association
• /
• v.57 no.2
• /
• pp.73-85
• /
• 2024

Predicting water quality of rivers and reservoirs is necessary for the management of water resources. Artificial Neural Networks (ANNs) have been used in many studies to predict water quality with high accuracy. Previous studies have used Gradient Descent (GD)-based optimizers as an optimizer, an operator of ANN that searches parameters. However, GD-based optimizers have the disadvantages of the possibility of local optimal convergence and absence of a solution storage and comparison structure. This study developed improved optimizers to overcome the disadvantages of GD-based optimizers. Proposed optimizers are optimizers that combine adaptive moments (Adam) and Nesterov-accelerated adaptive moments (Nadam), which have low learning errors among GD-based optimizers, with Harmony Search (HS) or Novel Self-adaptive Harmony Search (NSHS). To evaluate the performance of Long Short-Term Memory (LSTM) using improved optimizers, the water quality data from the Dasan water quality monitoring station were used for training and prediction. Comparing the learning results, Mean Squared Error (MSE) of LSTM using Nadam combined with NSHS (NadamNSHS) was the lowest at 0.002921. In addition, the prediction rankings according to MSE and R² for the four water quality indices for each optimizer were compared. Comparing the average of ranking for each optimizer, it was confirmed that LSTM using NadamNSHS was the highest at 2.25.

• Journal of the Korean Society of Radiology
• /
• v.85 no.4
• /
• pp.714-726
• /
• 2024

Researchers have developed various algorithms utilizing artificial intelligence (AI) to automatically and objectively diagnose patterns and extent of pulmonary emphysema or interstitial lung diseases on chest CT scans. Studies show that AI-based quantification of emphysema on chest CT scans reveals a connection between an increase in the relative percentage of emphysema and a decline in lung function. Notably, quantifying centrilobular emphysema has proven helpful in predicting clinical symptoms or mortality rates of chronic obstructive pulmonary disease. In the context of interstitial lung diseases, AI can classify the usual interstitial pneumonia pattern on CT scans into categories like normal, ground-glass opacity, reticular opacity, honeycombing, emphysema, and consolidation. This classification accuracy is comparable to chest radiologists (70%-80%). However, the results generated by AI are influenced by factors such as scan parameters, reconstruction algorithms, radiation doses, and the training data used to develop the AI. These limitations currently restrict the widespread adoption of AI for quantifying pulmonary emphysema and interstitial lung diseases in daily clinical practice. This paper will showcase the authors' experience using AI for diagnosing and quantifying emphysema and interstitial lung diseases through case studies. We will primarily focus on the advantages and limitations of AI for these two diseases.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.205-212
• /
• 2024

This study developed a deep learning model that distinguishes the front (with garnish) and the back (without garnish) surface of the dried semi-finished product (dried bukak) for screening operation before transfter the dried bukak to oil heater using robot's vacuum gripper. For deep learning model training and verification, RGB images for the front and back surfaces of 400 dry bukak that treated by data preproccessing were obtained. YOLO-v5 was used as a base structure of deep learning model. The area, surface information labeling, and data augmentation techniques were applied from the acquired image. Parameters including mAP, mIoU, accumulation, recall, decision, and F1-score were selected to evaluate the performance of the developed YOLO-v5 deep learning model-based surface detection model. The mAP and mIoU on the front surface were 0.98 and 0.96, respectively, and on the back surface, they were 1.00 and 0.95, respectively. The results of binary classification for the two front and back classes were average 98.5%, recall 98.3%, decision 98.6%, and F1-score 98.4%. As a result, the developed model can classify the surface information of the dried bukak using RGB images, and it can be used to develop a robot-automated system for the surface detection process of the dried bukak before deep frying.