Search

This paper proposes a method for forecasting surface solar irradiance (SSI), the most critical factor in solar photovoltaic (PV) power generation. The proposed method uses 16-channel data obtained by the GEO-KOMPSAT-2A (GK2A) satellite of South Korea as the main data for SSI forecasting. To determine feature variables related to SSI from the 16-channel data, the differences and ratios between the channels were utilized. Additionally, to consider the fundamental characteristics of SSI originating from the sun, solar geometry parameters, such as solar declination (SD), solar elevation angle (SEA), and extraterrestrial solar radiation (ESR), were used. Deep learning-based feature selection (Deep-FS) was employed to select appropriate feature variables that affect SSI from various feature variables extracted from the 16-channel data. Lastly, spatio-temporal deep learning models, such as convolutional neural network–long short-term memory (CNN-LSTM) and CNN–gated recurrent unit (CNN-GRU), which can simultaneously reflect temporal and spatial characteristics, were used to forecast SSI. Experiments conducted to verify the proposed method against conventional methods confirmed that the proposed method delivers superior SSI forecasting performance. [ABSTRACT FROM AUTHOR]

As online shopping is becoming mainstream, driven by the social impact of Coronavirus disease-2019 (COVID-19) as well as the development of Internet services, the demand for autonomous delivery mobile robots is rapidly increasing. This trend has brought the autonomous mobile robot market to a new turning point, with expectations that numerous mobile robots will be driving on roads with traffic. To achieve these expectations, autonomous mobile robots should precisely perceive the situation on roads with traffic. In this paper, we revisit and implement a real-time traffic light recognition system with a proposed lightweight state recognition network and ratio-preserving zero padding, which is a two-stage system consisting of a traffic light detection (TLD) module and a traffic light status recognition (TLSR) module. For the TLSR module, this work proposes a lightweight state recognition network with a small number of weight parameters, because the TLD module needs more weight parameters to find the exact location of traffic lights. Then, the proposed effective and lightweight network architecture is constructed by using skip connection, multifeature maps with different sizes, and kernels of appropriately tuned sizes. Therefore, the network has a negligible impact on the overall processing time and minimal weight parameters while maintaining high performance. We also propose to utilize a ratio-preserving zero padding method for data preprocessing for the TLSR module to enhance recognition accuracy. For the TLD module, extensive evaluations with varying input sizes and backbone network types are conducted, and then appropriate values for those factors are determined, which strikes a balance between detection performance and processing time. Finally, we demonstrate that our traffic light recognition system, utilizing the TLD module's determined parameters, the proposed network architecture for the TLSR module, and the ratio-preserving zero padding method can reliably detect the location and state of traffic lights in real-world videos recorded in Gumi and Deagu, Korea, while maintaining at least 30 frames per second for real-time operation. [ABSTRACT FROM AUTHOR]

The management of product quality is a crucial process in factory manufacturing. However, this approach still has some limitations, e.g., depending on the expertise of the engineer in evaluating products and being time consuming. Various approaches using deep learning in automatic defect detection and classification during production have been introduced to overcome these limitations. In this paper, we study applying different deep learning approaches and computer vision methods to detect scratches on the surface of microfasteners used in rechargeable batteries. Furthermore, we introduce an architecture with statistical quality control (SQC) to continuously improve the efficiency and accuracy of the product quality. The proposed architecture takes advantage of the capability of deep learning approaches, computer vision techniques, and SQC to automate the defect detection process and quality improvement. The proposed approach was evaluated using a real dataset comprising 1150 microfastener surface images obtained from a factory in Korea. In the study, we compared the direct and indirect prediction methods for predicting the scratches on the surface of the microfasteners and achieved the best accuracy of 0.91 with the indirect prediction approach. Notably, the indirect prediction method was more efficient than the traditional one. Furthermore, using control charts in SQC to analyze predicted defects in the production process helped operators understand the efficiency of the production line and make appropriate decisions in the manufacturing process, hence improving product quality management. [ABSTRACT FROM AUTHOR]

Federated Learning (FL) is an effective framework for a distributed system that constructs a powerful global deep learning model, which diminishes the local bias and accommodates the successful aggregation of locally trained models with heterogeneous datasets. However, when local datasets have the non-IID attribute, the optimization metric tends to diverge or show unstable convergence in the trajectory space. This paper delves into building a global model for the distributed Smart Grid environment, with regionally cumulated three solar energy datasets from January 2017 to August 2021 in a decentralized power grid in South Korea via FL. This distributed energy network involves local properties and physical distance between the regions, which raises a fundamental question of "Will time-serially curated non-IID local features be effective in constructing a global regression model?". This paper probes this question by leveraging FL and conducts the theoretically viable non-IID case-by-case convergence analysis, providing the interpretation of the embedded temporal non-IID features and application on real-world data. Moreover, most of the FL studies predetermine the global update period, which lacks applicability when adapting FL in actual practice. As FL is a cumulative-basis structure, the update term is a crucial factor that needs to be carefully selected. This paper articulates this problem and explores the effective update period via multiple experiments on the 4.5 years of solar energy dataset, and to the best of my knowledge, this is the first literature that presents the optimal update period in the FL regression in an energy domain. [ABSTRACT FROM AUTHOR]

The smart intersection (SI) systems, as they are named in the Republic of Korea, are part of the ITS services implemented under local government projects with financial support from the central government. They collect real-time traffic data available at signalized intersections with advanced detection systems for surveillance purposes only. A traffic signal method utilizing such valuable data has been desirable but unavailable as yet in practice. This paper proposes a new approach to designing traffic signal timings, reflecting the demand changing in real time, by utilizing SI surveillance data. The proposed artificial neural network model generates suitable traffic signal timings trained to be near optimum based on surveillance data for each directional movement. [ABSTRACT FROM AUTHOR]

Simple Summary: Over the years, the monitoring of cattle behavior has been recognized as an essential aspect of ensuring their health and welfare. In this study, we propose a framework based on artificial intelligence for monitoring the behavior of individual cattle through action recognition and tracking over time. Our research focuses specifically on studying the behavior of Hanwoo cattle, a native breed of Korea. To achieve this, we deployed a network of CCTV (closed-circuit television) cameras strategically placed within a closed farm, showcasing the effectiveness of non-intrusive sensors in capturing real-world information. Furthermore, we devised techniques to tackle challenges such as occlusion, size variations, and motion deformation. Our proposed technology represents a significant advancement in the field of precision livestock farming. By enabling the monitoring of individual animal behavior over time, it offers valuable insights that optimize the management of farms. This innovative approach enhances the efficiency and effectiveness of farm operations, ultimately contributing to the overall success and progress of the agriculture industry. Cattle behavior recognition is essential for monitoring their health and welfare. Existing techniques for behavior recognition in closed barns typically rely on direct observation to detect changes using wearable devices or surveillance cameras. While promising progress has been made in this field, monitoring individual cattle, especially those with similar visual characteristics, remains challenging due to numerous factors such as occlusion, scale variations, and pose changes. Accurate and consistent individual identification over time is therefore essential to overcome these challenges. To address this issue, this paper introduces an approach for multiview monitoring of individual cattle behavior based on action recognition using video data. The proposed system takes an image sequence as input and utilizes a detector to identify hierarchical actions categorized as part and individual actions. These regions of interest are then inputted into a tracking and identification mechanism, enabling the system to continuously track each individual in the scene and assign them a unique identification number. By implementing this approach, cattle behavior is continuously monitored, and statistical analysis is conducted to assess changes in behavior in the time domain. The effectiveness of the proposed framework is demonstrated through quantitative and qualitative experimental results obtained from our Hanwoo cattle video database. Overall, this study tackles the challenges encountered in real farm indoor scenarios, capturing spatiotemporal information and enabling automatic recognition of cattle behavior for precision livestock farming. [ABSTRACT FROM AUTHOR]

In predicting energy consumption, classic econometric and statistical models are used to forecast energy consumption. These models may have limitations in an increasingly fast-changing energy market, which requires big data analysis of energy consumption patterns and relevant variables using complex mathematical tools. In current literature, there are minimal comparison studies reviewing machine learning algorithms to predict energy consumption in Korea. To bridge this gap, this paper compared three different machine learning algorithms, namely the Random Forest (RF) model, XGBoost (XGB) model, and Long Short-Term Memory (LSTM) model. These algorithms were applied in Period 1 (prior to the onset of the COVID-19 pandemic) and Period 2 (after the onset of the COVID-19 pandemic). Period 1 was characterized by an upward trend in energy consumption, while Period 2 showed a reduction in energy consumption. LSTM performed best in its prediction power specifically in Period 1, and RF outperformed the other models in Period 2. Findings, therefore, suggested the applicability of machine learning to forecast energy consumption and also demonstrated that traditional econometric approaches may outperform machine learning when there is less unknown irregularity in the time series, but machine learning can work better with unexpected irregular time series data. [ABSTRACT FROM AUTHOR]

Weighting the Hanwoo (Korean cattle) is very important for Korean beef producers when selling the Hanwoo at the right time. Recently, research is being conducted on the automatic prediction of the weight of Hanwoo only through images with the achievement of research using deep learning and image recognition. In this paper, we propose a method for the automatic weight prediction of Hanwoo using the Bayesian ridge algorithm on RGB-D images. The proposed system consists of three parts: segmentation, extraction of features, and estimation of the weight of Korean cattle from a given RGB-D image. The first step is to segment the Hanwoo area from a given RGB-D image using depth information and color information, respectively, and then combine them to perform optimal segmentation. Additionally, we correct the posture using ellipse fitting on segmented body image. The second step is to extract features for weight prediction from the segmented Hanwoo image. We extracted three features: size, shape, and gradients. The third step is to find the optimal machine learning model by comparing eight types of well-known machine learning models. In this step, we compared each model with the aim of finding an efficient model that is lightweight and can be used in an embedded system in the real field. To evaluate the performance of the proposed weight prediction system, we collected 353 RGB-D images from livestock farms in Wonju, Gangwon-do in Korea. In the experimental results, random forest showed the best performance, and the Bayesian ridge model is the second best in MSE or the coefficient of determination. However, we suggest that the Bayesian ridge model is the most optimal model in the aspect of time complexity and space complexity. Finally, it is expected that the proposed system will be casually used to determine the shipping time of Hanwoo in wild farms for a portable commercial device. [ABSTRACT FROM AUTHOR]

Ginseng is an important medicinal plant in Korea. The roots of the ginseng plant have medicinal properties; thus, it is very important to maintain the quality of ginseng roots. Root rot disease is a major disease that affects the quality of ginseng roots. It is important to predict this disease before it causes severe damage to the plants. Hence, there is a need for a non-destructive method to identify root rot disease in ginseng plants. In this paper, a method to identify the root rot disease by analyzing the RGB plant images using image processing and deep learning is proposed. Initially, plant segmentation is performed, and then the noise regions are removed in the plant images. These images are given as input to the proposed linear deep learning model to identify root rot disease in ginseng plants. Transfer learning models are also applied to these images. The performance of the proposed method is promising in identifying root rot disease. [ABSTRACT FROM AUTHOR]

The 3D point cloud reconstruction from photos taken by an unmanned aerial vehicle (UAV) is a promising tool for monitoring and managing risks of cut-slopes. However, surface changes on cut-slopes are likely to be hidden by seasonal vegetation variations on the cut-slopes. This paper proposes a vegetation removal method for 3D reconstructed point clouds using (1) a 2D image segmentation deep learning model and (2) projection matrices available from photogrammetry. For a given point cloud, each 3D point of it is reprojected into the image coordinates by the projection matrices to determine if it belongs to vegetation or not using the 2D image segmentation model. The 3D points belonging to vegetation in the 2D images are deleted from the point cloud. The effort to build a 2D image segmentation model was significantly reduced by using U-Net with the dataset prepared by the colour index method complemented by manual trimming. The proposed method was applied to a cut-slope in Doam Dam in South Korea, and showed that vegetation from the two point clouds of the cut-slope at winter and summer was removed successfully. The M3C2 distance between the two vegetation-removed point clouds showed a feasibility of the proposed method as a tool to reveal actual change of cut-slopes without the effect of vegetation. [ABSTRACT FROM AUTHOR]

This paper is concerned with posture recognition using ensemble convolutional neural networks (CNNs) in home environments. With the increasing number of elderly people living alone at home, posture recognition is very important for helping elderly people cope with sudden danger. Traditionally, to recognize posture, it was necessary to obtain the coordinates of the body points, depth, frame information of video, and so on. In conventional machine learning, there is a limitation in recognizing posture directly using only an image. However, with advancements in the latest deep learning, it is possible to achieve good performance in posture recognition using only an image. Thus, we performed experiments based on VGGNet, ResNet, DenseNet, InceptionResNet, and Xception as pre-trained CNNs using five types of preprocessing. On the basis of these deep learning methods, we finally present the ensemble deep model combined by majority and average methods. The experiments were performed by a posture database constructed at the Electronics and Telecommunications Research Institute (ETRI), Korea. This database consists of 51,000 images with 10 postures from 51 home environments. The experimental results reveal that the ensemble system by InceptionResNetV2s with five types of preprocessing shows good performance in comparison to other combination methods and the pre-trained CNN itself. [ABSTRACT FROM AUTHOR]

We present SolarFlux Predictor, a novel deep-learning model designed to revolutionize photovoltaic (PV) power forecasting in South Korea. This model uses a self-attention-based temporal convolutional network (TCN) to process and predict PV outputs with high precision. We perform meticulous data preprocessing to ensure accurate data normalization and outlier rectification, which are vital for reliable PV power data analysis. The TCN layers are crucial for capturing temporal patterns in PV energy data; we complement them with the teacher forcing technique during the training phase to significantly enhance the sequence prediction accuracy. By optimizing hyperparameters with Optuna, we further improve the model's performance. Our model incorporates multi-head self-attention mechanisms to focus on the most impactful temporal features, thereby improving forecasting accuracy. In validations against datasets from nine regions in South Korea, SolarFlux outperformed conventional methods. The results indicate that SolarFlux is a robust tool for optimizing PV systems' management and operational efficiency and can contribute to South Korea's pursuit of sustainable energy solutions. [ABSTRACT FROM AUTHOR]

Urban transportation systems are increasingly burdened by traffic congestion, a consequence of population growth and heightened reliance on private vehicles. This congestion not only disrupts travel efficiency but also undermines productivity and urban resident's overall well-being. A critical step in addressing this challenge is the accurate prediction of bus travel times, which is essential for mitigating congestion and improving the experience of public transport users. To tackle this issue, this study introduces the Hybrid Temporal Forecasting Network (HTF-NET) model, a framework that integrates machine learning techniques. The model combines an attention mechanism with Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) layers, enhancing its predictive capabilities. Further refinement is achieved through a Support Vector Regressor (SVR), enabling the generation of precise bus travel time predictions. To evaluate the performance of the HTF-NET model, comparative analyses are conducted with six deep learning models using real-world digital tachograph (DTG) data obtained from intracity buses in Cheonan City, South Korea. These models includes various architectures, including different configurations of LSTM and GRU, such as bidirectional and stacked architectures. The primary focus of the study is on predicting travel times from the Namchang Village bus stop to the Dongnam-gu Public Health Center, a crucial route in the urban transport network. Various experimental scenarios are explored, incorporating overall test data, and weekday and weekend data, with and without weather information, and considering different route lengths. Comparative evaluations against a baseline ARIMA model underscore the performance of the HTF-NET model. Particularly noteworthy is the significant improvement in prediction accuracy achieved through the incorporation of weather data. Evaluation metrics, including root mean squared error (RMSE), mean absolute error (MAE), and mean squared error (MSE), consistently highlight the superiority of the HTF-NET model, outperforming the baseline ARIMA model by a margin of 63.27% in terms of the RMSE. These findings provide valuable insights for transit agencies and policymakers, facilitating informed decisions regarding the management and optimization of public transportation systems. [ABSTRACT FROM AUTHOR]

With the increasing number of social media users in recent years, news in various fields, such as politics, economics, and so on, can be easily accessed by users. However, most news spread through social networks including Twitter, Facebook, and Instagram has unknown sources, thus having a significant impact on news consumers. Fake news on COVID-19, which is affecting the global population, is propagating quickly and causes social disorder. Thus, a lot of research is being conducted on the detection of fake news on COVID-19 but is facing the problem of a lack of datasets. In order to alleviate the problem, we built a dataset on COVID-19 fake news from fact-checking websites in Korea and propose deep learning for detecting fake news on COVID-19 using the datasets. The proposed model is pre-trained with large-scale data and then performs transfer learning through a BiLSTM model. Moreover, we propose a method for initializing the hidden and cell states of the BiLSTM model to a [CLS] token instead of a zero vector. Through experiments, the proposed model showed that the accuracy is 78.8%, which was improved by 8% compared with the linear model as a baseline model, and that transfer learning can be useful with a small amount of data as we know it. A [CLS] token containing sentence information as the initial state of the BiLSTM can contribute to a performance improvement in the model. [ABSTRACT FROM AUTHOR]

The most important thing to operate a power system is that the power supply should be close to the power demand. In order to predict the amount of electric power transaction (EPT), it is important to choose and decide the variable and its starting date. In this paper, variables that could be acquired one the starting day of prediction were chosen. This paper designated date, temperature and special day as variables to predict the amount of EPT of the Korea Electric Power company. This paper also used temperature data from a year ago to predict the next year. To do this, we proposed single deep learning algorithms and hybrid deep learning algorithms. The former included multi-layer perceptron (MLP), convolution neural network (CNN), long short-term memory (LSTM), gated recurrent unit (GRU), support vector machine regression (SVR), and adaptive network-based fuzzy inference system (ANFIS). The latter included LSTM + CNN and CNN + LSTM. We then confirmed the improvement of accuracy for prediction using pre-processed variables compared to original variables We also assigned two years of test data during 2017–2018 as variable data to measure high prediction accuracy. We then selected a high-accuracy algorithm after measuring root mean square error (RMSE) and mean absolute percent error (MAPE). Finally, we predicted the amount of EPT in 2018 and then measured the error for each proposed algorithm. With these acquired error data, we obtained a model for predicting the amount of EPT with a high accuracy. [ABSTRACT FROM AUTHOR]

In 2020 and 2021, humanity lived in fear due to the COVID-19 pandemic. However, with the development of artificial intelligence technology, mankind is attempting to tackle many challenges from currently unpredictable epidemics. Korean society has been exposed to various infectious diseases since the Korean War in 1950, and to overcome them, the six most serious cases in National Notifiable Infectious Diseases (NNIDs) category I were defined. Although most infectious diseases have been overcome, viral hepatitis A has been on the rise in Korean society since 2010. Therefore, in this paper, the prediction of viral hepatitis A, which is rapidly spreading in Korean society, was predicted by region using the deep learning technique and a publicly available dataset. For this study, we gathered information from five organizations based on the open data policy: Korea Centers for Disease Control and Prevention (KCDC), National Institute of Environmental Research (NIER), Korea Meteorological Agency (KMA), Public Open Data Portal, and Korea Environment Corporation (KECO). Patient information, water environment information, weather information, population information, and air pollution information were acquired and correlations were identified. Next, an epidemic outbreak prediction was performed using data preprocessing and 3D LSTM. The experimental results were compared with various machine learning methods through RMSE. In this paper, we attempted to predict regional epidemic outbreaks of hepatitis A by linking the open data environment with deep learning. It is expected that the experimental process and results will be used to present the importance and usefulness of establishing an open data environment. [ABSTRACT FROM AUTHOR]

Several cases of Industrial Internet of Things (IIoT) attacks with zero-day vulnerabilities have been reported. To prevent these attacks, it is necessary to apply an abnormal behavior detection method; however, there are three main problems that make it hard. First, there are various industrial communication protocols. Instead of IT environments, many unstandardized protocols, which are usually defined by vendors, are used. Second, legacy devices are commonly used, not only EOS (End-of-service), but also EoL (End-of-Life). And last, the analysis of collected data is necessary for defining normal behavior. This behavior should be separately defined in each IIoT. Therefore, it is difficult to apply abnormal behavior detection in environments where economic and human investment is difficult. To solve these problems, we propose a deep learning based abnormal behavior detection technique that utilizes IIoT communication patterns. The proposed method uses a deep learning technique to train periodic data acquisition sequences, which is one of the common characteristics of IIoT. The trained model determined the sequence of packet is normal. The proposed technique can be applied without an additional analysis. The proposed method is expected to prevent security threats by proactively detecting cyberattacks. To verify the proposed method, a dataset was collected from the Korea Electric Power Control System. The model that defines normal behavior based on the application layer exhibits an accuracy of 79.6%. The other model, defining normal behavior based on the transport layer, has an accuracy of 80.9%. In these two models, most false positives and false negatives only occur when the abnormal packet is in a sequence. [ABSTRACT FROM AUTHOR]

In current railway rails, trains are propelled by the rolling contact between iron wheels and iron rails, and the high frequency of train repetition on rails results in a significant load exertion on a very small area where the wheel and rail come into contact. Furthermore, a contact stress beyond the allowable stress of the rail may lead to cracks due to plastic deformation. The railway rail, which is the primary contact surface between the wheel and the rail, is prone to rolling contact fatigue cracks. Therefore, a thorough inspection and diagnosis of the condition of the cracks is necessary to prevent fracture. The Detailed Guideline on the Performance Evaluation of Track Facilities in South Korea specifies the detailed requirements for the methods and procedures for conducting track performance evaluations. However, diagnosing rail surface damage and determining the severity solely rely on visual inspection, which depends on the qualitative evaluation and subjective judgment of the inspector. Against this backdrop, rail surface defect detection was investigated using Fast R-CNN in this study. To test the feasibility of the model, we constructed a dataset of rail surface defect images. Through field investigation, 1300 images of rail surface defects were obtained. Aged rails collected from the field were processed, and 1300 images of internal defects were generated through SEM testing; therefore, a total of 1300 pieces of learning data were constructed. The detection results indicated that the mean average precision was 94.9%. The Fast R-CNN exhibited high efficiency in detecting rail surface defects, and it demonstrated a superior recognition performance compared with other algorithms. [ABSTRACT FROM AUTHOR]

The accurate detection of P-wave FAP (First-Arrival Picking) in seismic signals is crucial across various industrial domains, including coal and oil exploration, tunnel construction, hydraulic fracturing, and earthquake early warning systems. At present, P-wave FAP detection relies on manual identification by experts and automated methods using Short-Term Average to Long-Term Average algorithms. However, these approaches encounter significant performance challenges, especially in the presence of real-time background noise. To overcome this limitation, this study proposes a novel P-wave FAP detection method that employs the U-Net model and incorporates spectrogram transformation techniques for seismic signals. Seismic signals, similar to those encountered in South Korea, were generated using the stochastic model simulation program. Synthesized WGN (White Gaussian Noise) was added to replicate background noise. The resulting signals were transformed into 2D spectrogram images and used as input data for the U-Net model, ensuring precise P-wave FAP detection. In the experimental result, it demonstrated strong performance metrics, achieving an MSE of 0.0031 and an MAE of 0.0177, and an RMSE of 0.0195. Additionally, it exhibited precise FAP detection capabilities in image prediction. The developed U-Net-based model exhibited exceptional performance in accurately detecting P-wave FAP in seismic signals with varying amplitudes. Through the developed model, we aim to contribute to the advancement of microseismic monitoring technology used in various industrial fields. [ABSTRACT FROM AUTHOR]

Multiscale forecasting of groundwater levels (GWLs) is essential for ensuring the sustainable management of groundwater resources, particularly considering the potential impacts of climate change. Such forecasting requires a model that is not only accurate in predicting GWLs but also computationally efficient, ensuring its suitability for practical applications. In this study, a temporal convolutional network (TCN) is implemented to forecast GWLs for 17 monitoring wells possessing diverse hydrogeological characteristics, located across South Korea. Using deep learning, the influence of meteorological variables (i.e., temperature, precipitation) on the forecasted GWLs was investigated by dividing the input features into three categories. Additionally, the models were developed for three forecast intervals (at 1-, 3-, and 6-month lead times) using each category input. When compared with state-of-the-art models, that is, long short-term memory (LSTM) and artificial neural network (ANN), the TCN model showed superior performance and required much less computational complexity. On average, the TCN model outperformed the LSTM model by 24%, 21%, and 25%, and the ANN model by 24%, 37%, and 47%, respectively, for 1-, 3-, and 6-month lead times. Based on these results, the proposed TCN model can be used for real-time GWL forecasting in hydrological applications. [ABSTRACT FROM AUTHOR]

The recent large-scale fire incidents on construction sites in South Korea have highlighted the need for computer vision technology to detect fire risks before an actual occurrence of fire. This study developed a proactive fire risk detection system by detecting the coexistence of an ignition source (sparks) and a combustible material (urethane foam or Styrofoam) using object detection on images from a surveillance camera. Statistical analysis was carried out on fire incidences on construction sites in South Korea to provide insight into the cause of the large-scale fire incidents. Labeling approaches were discussed to improve the performance of the object detectors for sparks and urethane foams. Detecting ignition sources and combustible materials at a distance was discussed in order to improve the performance for long-distance objects. Two candidate deep learning models, Yolov5 and EfficientDet, were compared in their performance. It was found that Yolov5 showed slightly higher mAP performances: Yolov5 models showed mAPs from 87% to 90% and EfficientDet models showed mAPs from 82% to 87%, depending on the complexity of the model. However, Yolov5 showed distinctive advantages over EfficientDet in terms of easiness and speed of learning. [ABSTRACT FROM AUTHOR]

Simple Summary: Over the years, the identification of individual cattle has assumed a pivotal role in health monitoring, reproduction management, behavioral research, and performance tracking. In this study, we propose a method based on artificial intelligence for identifying known and new (unknown) individual Hanwoo cattle, a native breed of Korea, by utilizing cattle's face images. To accomplish this, we strategically positioned a network of CCTV cameras within a closed farm, demonstrating the efficacy of non-intrusive sensors in capturing real-world data. Furthermore, we devised open-set techniques to tackle challenges such as varying illumination, overlapping objects, and fluctuations in cattle's face orientations. Our research method not only demonstrated excellent recognition performance in complex real-world cattle's datasets, but can also be applied to open-set scenarios, wherein unmarked or new cattle may join the herd. Our proposed method can be readily adapted to identifying various livestock species, offering real-time individual recognition, which yields valuable insights for farm management. This deep learning approach amplifies the efficiency of farm operations, thus playing a pivotal role in advancing the agriculture industry as a whole. Accurate identification of individual cattle is of paramount importance in precision livestock farming, enabling the monitoring of cattle behavior, disease prevention, and enhanced animal welfare. Unlike human faces, the faces of most Hanwoo cattle, a native breed of Korea, exhibit significant similarities and have the same body color, posing a substantial challenge in accurately distinguishing between individual cattle. In this study, we sought to extend the closed-set scope (only including identifying known individuals) to a more-adaptable open-set recognition scenario (identifying both known and unknown individuals) termed Cattle's Face Open-Set Recognition (CFOSR). By integrating open-set techniques to enhance the closed-set accuracy, the proposed method simultaneously addresses the open-set scenario. In CFOSR, the objective is to develop a trained model capable of accurately identifying known individuals, while effectively handling unknown or novel individuals, even in cases where the model has been trained solely on known individuals. To address this challenge, we propose a novel approach that integrates Adversarial Reciprocal Points Learning (ARPL), a state-of-the-art open-set recognition method, with the effectiveness of Additive Margin Softmax loss (AM-Softmax). ARPL was leveraged to mitigate the overlap between spaces of known and unknown or unregistered cattle. At the same time, AM-Softmax was chosen over the conventional Cross-Entropy loss (CE) to classify known individuals. The empirical results obtained from a real-world dataset demonstrated the effectiveness of the ARPL and AM-Softmax techniques in achieving both intra-class compactness and inter-class separability. Notably, the results of the open-set recognition and closed-set recognition validated the superior performance of our proposed method compared to existing algorithms. To be more precise, our method achieved an AUROC of 91.84 and an OSCR of 87.85 in the context of open-set recognition on a complex dataset. Simultaneously, it demonstrated an accuracy of 94.46 for closed-set recognition. We believe that our study provides a novel vision to improve the classification accuracy of the closed set. Simultaneously, it holds the potential to significantly contribute to herd monitoring and inventory management, especially in scenarios involving the presence of unknown or novel cattle. [ABSTRACT FROM AUTHOR]

As the number of researchers in South Korea has grown, there is increasing dissatisfaction with the selection process for national research and development (R&D) projects among unsuccessful applicants. In this study, we designed a system that can recommend the best possible R&D evaluators using big data that are collected from related systems, refined, and analyzed. Our big data recommendation system compares keywords extracted from applications and from the full-text of the achievements of the evaluator candidates. Weights for different keywords are scored using the term frequency–inverse document frequency algorithm. Comparing the keywords extracted from the achievement of the evaluator candidates', a project comparison module searches, scores, and ranks these achievements similarly to the project applications. The similarity scoring module calculates the overall similarity scores for different candidates based on the project comparison module scores. To assess the performance of the evaluator candidate recommendation system, 61 applications in three Review Board (RB) research fields (system fusion, organic biochemistry, and Korean literature) were recommended as the evaluator candidates by the recommendation system in the same manner as the RB's recommendation. Our tests reveal that the evaluator candidates recommended by the Korean Review Board and those recommended by our system for 61 applications in different areas, were the same. However, our system performed the recommendation in less time with no bias and fewer personnel. The system requiresrevisions to reflect qualitative indicators, such as journal reputation, before it can entirely replace the current evaluator recommendation process. [ABSTRACT FROM AUTHOR]

Coronavirus 2019 (COVID-19) has posed a serious threat to the lives and health of the majority of people worldwide. Since the early days of the outbreak, South Korea's government and citizens have made persistent efforts to provide effective prevention against further spread of the disease. In particular, the participation of individual citizens in complying with the necessary code of conduct to prevent spread of the infection, through measures such as social distancing and mask wearing, is as instrumental as the geographical tracking of the trajectory of the infected. In this paper, we propose an activity recognition method based on a wristband equipped with an IR array and inertial measurement unit (IMU) to detect individual compliance with codes of personal hygiene management, such as mask wearing, which are recommended to prevent the spread of infectious diseases. The results of activity recognition were comparatively analyzed by applying conventional machine learning algorithms and convolutional neural networks (CNNs) to the IMU time series and IR array thermal images collected from 25 subjects. When CNN and 24 × 32 thermal images were used, 97.8% accuracy was achieved (best performance), and when 6 × 8 low-resolution thermal images were used, similar performance with 97.1% accuracy was obtained. In the case of using IMU, the performance of activity recognition was lower than that obtained with the IR array, but an accuracy of 93% was achieved even in the case of applying machine learning algorithms, indicating that it is more suitable for wearable devices with low computational capability. [ABSTRACT FROM AUTHOR]

Worldwide COVID-19 infections have caused various problems throughout different countries. In the case of Korea, problems related to the demand for medical care concerning wards and doctors are serious, which were already slowly worsening problems in Korea before the COVID-19 pandemic. In this paper, we propose the direction of developing a system by combining artificial intelligence technology with limited areas that do not require high expertise in the rehabilitation medical field that should be improved in Korea through the prevention of bedsores and leg rehabilitation methods. Regarding the introduction of artificial intelligence technology, medical and related laws and regulations were quite limited, so the actual needs of domestic rehabilitation doctors and advice on the hospital environment were obtained. Satisfaction with the test content was high, the degree of provision of important medical data was 95%, and the angular error was within 5 degrees and suitable for recovery confirmation. [ABSTRACT FROM AUTHOR]

We propose a novel system leveraging deep learning-based methods to predict urban traffic accidents and estimate their severity. The major challenge is the data imbalance problem in traffic accident prediction. The problem is caused by numerous zero values in the dataset due to the rarity of traffic accidents. To address the issue, we propose a grid-clustered feature map with the ideas of grids and cells. To predict the occurrence of accidents in the grid, we introduce an accident detector that combines the power of a Convolutional Neural Network (CNN) with a Deep Neural Network (DNN). Then, hierarchical DNNs are supposed to be an accident risk classifier to estimate the risk of each cell in the accident-occurrence grid. The proposed system can effectively reduce instances with no traffic accidents. Furthermore, we introduce the concept of the Accident Risk Index (ARI) to better represent the severity of risk at each cell. Also, we consider all the explanatory variables, such as dangerous driving behaviors, traffic mobility, and safety facility information, that can be related to traffic accidents. To improve the prediction accuracy, we further take into consideration all the explanatory variables, such as dangerous driving behaviors, traffic mobility, and safety facility information, that can be related to traffic accidents. In the experiment, we highlight the benefits of our method for urban traffic accident management by significantly improving model performance compared to the baselines. The feasibility and applicability of the proposed system are validated in the data of Daejeon City, Republic of Korea. The proposed prediction system can dynamically advise and recommend commuters, traffic management systems, and city planners on alternatives, optimizations, and interventions. [ABSTRACT FROM AUTHOR]

Early and accurate flood forecasting and warning for urban flood risk areas is an essential factor to reduce flood damage. This paper presents the urban flood forecasting and warning process to reduce damage in the main flood risk area of South Korea. This process is developed based on the rainfall-runoff model and deep learning model. A model-driven method was devised to construct the accurate physical model with combined inland-river and flood control facilities, such as pump stations and underground storages. To calibrate the rainfall-runoff model, data of gauging stations and pump stations of an urban stream in August 2020 were used, and the model result was presented as an R2 value of 0.63~0.79. Accurate flood warning criteria of the urban stream were analyzed according to the various rainfall scenarios from the model-driven method. As flood forecasting and warning in the urban stream, deep learning models, vanilla ANN, Long Short-Term Memory (LSTM), Stack-LSTM, and Bidirectional LSTM were constructed. Deep learning models using 10-min hydrological time-series data from gauging stations were trained to warn of expected flood risks based on the water level in the urban stream. A forecasting and warning method that applied the bidirectional LSTM showed an R2 value of 0.9 for the water level forecast with 30 min lead time, indicating the possibility of effective flood forecasting and warning. This case study aims to contribute to the reduction of casualties and flood damage in urban streams and accurate flood warnings in typical urban flood risk areas of South Korea. The developed urban flood forecasting and warning process can be applied effectively as a non-structural measure to mitigate urban flood damage and can be extended considering watershed characteristics. [ABSTRACT FROM AUTHOR]

Air pollution is a critical problem that is of major concern worldwide. South Korea is one of the countries most affected by air pollution. Rapid urbanization and industrialization in South Korea have induced air pollution in multiple forms, such as smoke from factories and exhaust from vehicles. In this paper, we perform a comparative analysis of predictive models for fine particulate matter in Daejeon, the fifth largest city in South Korea. This study is conducted for three purposes. The first purpose is to determine the factors that may cause air pollution. Two main factors are considered: meteorological and traffic. The second purpose is to find an optimal predictive model for air pollutant concentration. We apply machine learning and deep learning models to the collected dataset to predict hourly air pollutant concentrations. The accuracy of the deep learning models is better than that of the machine learning models. The third purpose is to analyze the influence of road conditions on predicting air pollutant concentration. Experimental results demonstrate that considering wind direction and wind speed could significantly decrease the error rate of the predictive models. [ABSTRACT FROM AUTHOR]

The energy manufacturers are required to produce an accurate amount of energy by meeting the energy requirements at the end-user side. Consequently, energy prediction becomes an essential role in the electric industrial zone. In this paper, we propose the hybrid ensemble deep learning model, which combines multilayer perceptron (MLP), convolutional neural network (CNN), long short-term memory (LSTM), and hybrid CNN-LSTM to improve the forecasting performance. These DL architectures are more popular and better than other machine learning (ML) models for time series electrical load prediction. Therefore, hourly-based energy data are collected from Jeju Island, South Korea, and applied for forecasting. We considered external features associated with meteorological conditions affecting energy. Two-year training and one-year testing data are preprocessed and arranged to reform the times series, which are then trained in each DL model. The forecasting results of the proposed ensemble model are evaluated by using mean square error (MSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). Error metrics are compared with DL stand-alone models such as MLP, CNN, LSTM, and CNN-LSTM. Our ensemble model provides better performance than other forecasting models, providing minimum MAPE at 0.75%, and was proven to be inherently symmetric for forecasting time-series energy and demand data, which is of utmost concern to the power system sector. [ABSTRACT FROM AUTHOR]

COVID-19 has further aggravated problems by compelling people to stay indoors and limit social interactions, leading to a worsening of the depression situation. This study aimed to construct a TabNet model combined with SHapley Additive exPlanations (SHAP) to predict depression in South Korean society during the COVID-19 pandemic. We used a tabular dataset extracted from the Seoul Welfare Survey with a total of 3027 samples. The TabNet model was trained on this dataset, and its performance was compared to that of several other machine learning models, including Random Forest, eXtreme Gradient Boosting, Light Gradient Boosting, and CatBoost. According to the results, the TabNet model achieved an Area under the receiver operating characteristic curve value (AUC) of 0.9957 on the training set and an AUC of 0.9937 on the test set. Additionally, the study investigated the TabNet model's local interpretability using SHapley Additive exPlanations (SHAP) to provide post hoc global and local explanations for the proposed model. By combining the TabNet model with SHAP, our proposed model might offer a valuable tool for professionals in social fields, and psychologists without expert knowledge in the field of data analysis can easily comprehend the decision-making process of this AI model. [ABSTRACT FROM AUTHOR]

The Web has provided an increasing proportion of use as a medium for e-commerce in addition to various recommender systems. It can be used for analyzing recommendation system-based feedback (e.g., a form in which a user inputs their preferences for various items as numerical values into a specific evaluation system) to estimate customer interest; in addition, analyzing multi-modal types of feedback (e.g., product purchase traces, inquiry lists, inquiry times, and comments) with deep learning can also be used to estimate user interest. As many companies around the world promote their products through micro-influencers on the Web, related research has continued to predict the purchase conversion rate of the influencer through a variety of technologies. In this work, we present a multi-modal micro-influencer analysis scheme for a marketing maximization strategy. Our scheme uses the multi-modal data stored in Mecha Solution's own shopping mall of Korea, as well as famous Korean Internet platforms, and Coupang, Naver, and Oliveyoung's data such as article posting comments and statistics information. By extracting the main characteristics of the real article postings from real users as opposed to those from factitious influencers posting articles and comments and identifying articles other than advertisements, influencer scores are obtained, assuming that articles other than advertisements can further increase the purchase conversion rate. Based on influencer score, we propose a multi-modal micro-influencer analysis scheme that recommends influencers use content-based collaborative filtering and user-based collaborative filtering for items that the influencer has not yet reviewed. The experiment was implemented to prove that the proposed scheme successfully achieves this goal. [ABSTRACT FROM AUTHOR]

Simple Summary: Conventional manual counting methods for the monitoring of mosquito species and populations can hinder the accurate determination of the optimal timing for pest control in the field. In this study, a deep learning-based automated image analysis method was developed for the classification of eleven species of mosquito. The combination of color and fluorescence images enhanced the performance for live mosquito classification. The classification result of a 97.1% F1-score has demonstrated the potential of using an automatic measurement of mosquito species and populations in the field. The proposed technique could be adapted for establishing a mosquito monitoring and management system, which may contribute to preemptive quarantine and a reduction in the exposure to vector-borne diseases. Mosquitoes are one of the deadliest insects, causing harm to humans worldwide. Preemptive prevention and forecasting are important to prevent mosquito-borne diseases. However, current mosquito identification is mostly conducted manually, which consumes time, wastes labor, and causes human error. In this study, we developed an automatic image analysis method to identify mosquito species using a deep learning-based object detection technique. Color and fluorescence images of live mosquitoes were acquired using a mosquito capture device and were used to develop a deep learning-based object detection model. Among the deep learning-based object identification models, the combination of a swine transformer and a faster region-convolutional neural network model demonstrated the best performance, with a 91.7% F1-score. This indicates that the proposed automatic identification method can be rapidly applied for efficient analysis of species and populations of vector-borne mosquitoes with reduced labor in the field. [ABSTRACT FROM AUTHOR]

Clays in fault zones have low electrical resistivity, making electrical resistivity tomography (ERT) effective for fault investigations. However, traditional ERT inversion methods struggle to find a unique solution and produce unstable results owing to the ill-posed nature of the problem. To address this, a workflow integrating deep-learning (DL) technology with traditional ERT inversion is proposed. First, a deep-learning model named DL-ERT inversion that maps apparent resistivity data to subsurface resistivity models is developed. To create target-oriented training data, we use approximately 150 field borehole data acquired from various survey areas in South Korea. The DL-ERT inversion algorithm is based on a U-Net structure and includes an additional network called the borehole mixer to incorporate borehole information when available. The DL-ERT inversion model is trained in three stages: base model training, borehole mixer training, and fine-tuning. Results showed that the fine-tuning model provided the highest prediction accuracy for all test datasets. Next, the prediction of the trained model is used as the initial model for the deterministic inversion method to predict the final subsurface model. The efficiency and accuracy of the proposed workflow are demonstrated in fault detection using a field data example compared with traditional deterministic inversion. [ABSTRACT FROM AUTHOR]

A deep learning methodology was utilized to predict the bulb weights of garlic and onions in the Jeolla Province of Korea. The Korea Rural Economic Institute (KREI) operates the Outlook & Agricultural Statistics Information System (OASIS) platform, which provides actual measurements of garlic and onions. We trained the Neural Prophet (NP) lagged time-series model using this data. The NP model effectively handles lagged variables and their covariates by inserting a hidden layer. Our results indicate that the NP model performed with around 5% mean absolute error in predicting bulb weights, with a gap of 3.3 g and 4.7 g with average weights of 63.7 g and 129.9 g for garlic and onions, respectively. This experimental research was based on only three years of measurement data. Hence, the gap between observed and predicted data can be reduced by accumulating more measurement data in the future. [ABSTRACT FROM AUTHOR]

The strawberry market in South Korea is actually the largest market among horticultural crops. Strawberry cultivation in South Korea changed from field cultivation to facility cultivation in order to increase production. However, the decrease in production manpower due to aging is increasing the demand for the automation of strawberry cultivation. Predicting the harvest of strawberries is an important research topic, as strawberry production requires the most manpower for harvest. In addition, the growing environment has a great influence on strawberry production as hydroponic cultivation of strawberries is increasing. In this paper, we design and implement an integrated system that monitors strawberry hydroponic environmental data and determines when to harvest with the concept of IoT-Edge-AI-Cloud. The proposed monitoring system collects, stores and visualizes strawberry growing environment data. The proposed harvest decision system classifies the strawberry maturity level in images using a deep learning algorithm. The monitoring and analysis results are visualized in an integrated interface, which provides a variety of basic data for strawberry cultivation. Even if the strawberry cultivation area increases, the proposed system can be easily expanded and flexibly based on a virtualized container with the concept of IoT-Edge-AI-Cloud. The monitoring system was verified by monitoring a hydroponic strawberry environment for 4 months. In addition, the harvest decision system was verified using strawberry pictures acquired from Smart Berry Farm. [ABSTRACT FROM AUTHOR]

Due to the availability of smart metering infrastructure, high-resolution electric consumption data is readily available to study the dynamics of residential electric consumption at finely resolved spatial and temporal scales. Analyzing the electric consumption data enables the policymakers and building owners to understand consumer's demand-consumption behaviors. Furthermore, analysis and accurate forecasting of electric consumption are substantial for consumer involvement in time-of-use tariffs, critical peak pricing, and consumer-specific demand response initiatives. Alongside its vast economic and sustainability implications, such as energy wastage and decarbonization of the energy sector, accurate consumption forecasting facilitates power system planning and stable grid operations. Energy consumption forecasting is an active research area; despite the abundance of devised models, electric consumption forecasting in residential buildings remains challenging due to high occupant energy use behavior variability. Hence the search for an appropriate model for accurate electric consumption forecasting is ever continuing. To this aim, this paper presents a spatial and temporal ensemble forecasting model for short-term electric consumption forecasting. The proposed work involves exploring electric consumption profiles at the apartment level through cluster analysis based on the k-means algorithm. The ensemble forecasting model consists of two deep learning models; Long Short-Term Memory Unit (LSTM) and Gated Recurrent Unit (GRU). First, the apartment-level historical electric consumption data is clustered. Later the clusters are aggregated based on consumption profiles of consumers. At the building and floor level, the ensemble models are trained using aggregated electric consumption data. The proposed ensemble model forecasts the electric consumption at three spatial scales apartment, building, and floor level for hourly, daily, and weekly forecasting horizon. Furthermore, the impact of spatial-temporal granularity and cluster analysis on the prediction accuracy is analyzed. The dataset used in this study comprises high-resolution electric consumption data acquired through smart meters recorded on an hourly basis over the period of one year. The consumption data belongs to four multifamily residential buildings situated in an urban area of South Korea. To prove the effectiveness of our proposed forecasting model, we compared our model with widely known machine learning models and deep learning variants. The results achieved by our proposed ensemble scheme verify that model has learned the sequential behavior of electric consumption by producing superior performance with the lowest MAPE of 4.182 and 4.54 at building and floor level prediction, respectively. The experimental findings suggest that the model has efficiently captured the dynamic electric consumption characteristics to exploit ensemble model diversities and achieved lower forecasting error. The proposed ensemble forecasting scheme is well suited for predictive modeling and short-term load forecasting. [ABSTRACT FROM AUTHOR]

With the development of modern power systems (smart grid), energy consumption prediction becomes an essential aspect of resource planning and operations. In the last few decades, industrial and commercial buildings have thoroughly been investigated for consumption patterns. However, due to the unavailability of data, the residential buildings could not get much attention. During the last few years, many solutions have been devised for predicting electric consumption; however, it remains a challenging task due to the dynamic nature of residential consumption patterns. Therefore, a more robust solution is required to improve the model performance and achieve a better prediction accuracy. This paper presents an ensemble approach based on learning to a statistical model to predict the short-term energy consumption of a multifamily residential building. Our proposed approach utilizes Long Short-Term Memory (LSTM) and Kalman Filter (KF) to build an ensemble prediction model to predict short term energy demands of multifamily residential buildings. The proposed approach uses real energy data acquired from the multifamily residential building, South Korea. Different statistical measures are used, such as mean absolute error (MAE), root mean square error (RMSE), mean absolute percentage error (MAPE), and R 2 score, to evaluate the performance of the proposed approach and compare it with existing models. The experimental results reveal that the proposed approach predicts accurately and outperforms the existing models. Furthermore, a comparative analysis is performed to evaluate and compare the proposed model with conventional machine learning models. The experimental results show the effectiveness and significance of the proposed approach compared to existing energy prediction models. The proposed approach will support energy management to effectively plan and manage the energy supply and demands of multifamily residential buildings. [ABSTRACT FROM AUTHOR]

Automatic age estimation using panoramic dental radiographic images is an important procedure for forensics and personal oral healthcare. The accuracies of the age estimation have increased recently with the advances in deep neural networks (DNN), but DNN requires large sizes of the labeled dataset which is not always available. This study examined whether a deep neural network is able to estimate tooth ages when precise age information is not given. A deep neural network model was developed and applied to age estimation using an image augmentation technique. A total of 10,023 original images were classified according to age groups (in decades, from the 10s to the 70s). The proposed model was validated using a 10-fold cross-validation technique for precise evaluation, and the accuracies of the predicted tooth ages were calculated by varying the tolerance. The accuracies were 53.846% with a tolerance of ±5 years, 95.121% with ±15 years, and 99.581% with ±25 years, which means the probability for the estimation error to be larger than one age group is 0.419%. The results indicate that artificial intelligence has potential not only in the forensic aspect but also in the clinical aspect of oral care. [ABSTRACT FROM AUTHOR]

Authorities and policymakers in Korea have recently prioritized improving fire prevention and emergency response. Governments seek to enhance community safety for residents by constructing automated fire detection and identification systems. This study examined the efficacy of YOLOv6, a system for object identification running on an NVIDIA GPU platform, to identify fire-related items. Using metrics such as object identification speed, accuracy research, and time-sensitive real-world applications, we analyzed the influence of YOLOv6 on fire detection and identification efforts in Korea. We conducted trials using a fire dataset comprising 4000 photos collected through Google, YouTube, and other resources to evaluate the viability of YOLOv6 in fire recognition and detection tasks. According to the findings, YOLOv6's object identification performance was 0.98, with a typical recall of 0.96 and a precision of 0.83. The system achieved an MAE of 0.302%. These findings suggest that YOLOv6 is an effective technique for detecting and identifying fire-related items in photos in Korea. Multi-class object recognition using random forests, k-nearest neighbors, support vector, logistic regression, naive Bayes, and XGBoost was performed on the SFSC data to evaluate the system's capacity to identify fire-related objects. The results demonstrate that for fire-related objects, XGBoost achieved the highest object identification accuracy, with values of 0.717 and 0.767. This was followed by random forest, with values of 0.468 and 0.510. Finally, we tested YOLOv6 in a simulated fire evacuation scenario to gauge its practicality in emergencies. The results show that YOLOv6 can accurately identify fire-related items in real time within a response time of 0.66 s. Therefore, YOLOv6 is a viable option for fire detection and recognition in Korea. The XGBoost classifier provides the highest accuracy when attempting to identify objects, achieving remarkable results. Furthermore, the system accurately identifies fire-related objects while they are being detected in real-time. This makes YOLOv6 an effective tool to use in fire detection and identification initiatives. [ABSTRACT FROM AUTHOR]

Owing to the recent increase in mobile banking customers, studies exploring self-service channels and customer responses as distribution channels in the retail banking industry are also rapidly expanding. Moreover, with the emergence of big data and a series of artificial intelligence (AI) technologies, customer pattern analysis using deep learning, insurance fraud prevention, software development and various types of blockchain-based FinTech technologies, offline banks are disappearing. Accordingly, many commercial banks are attempting to find technological alternatives. However, maintaining a profitable bank branch is a crucial factor in the relationship between service quality and customer satisfaction because excellent service quality prevents existing customers from leaving. This study sought to quantitatively prove the change in customer service quality and profit due to the introduction of technology in the financial industry. We microscopically compared the effects between bank branch closures and changes in net profit using a time-series analysis. Specifically, we quantitatively analyzed actual customer attrition behavior with a time-series analysis across the three quarters before and after the closure of 88 branches of major commercial banks in South Korea in the Seoul metropolitan area and nearby cities. The findings prove that branch closures and multi-channel effects in the financial sector are gradually being resolved through immediate technology acceptance, contrary to popular concern. [ABSTRACT FROM AUTHOR]

The proportion of the inventory range associated with spare parts is often considered in the industrial context. Therefore, even minor improvements in forecasting the demand for spare parts can lead to substantial cost savings. Despite notable research efforts, demand forecasting remains challenging, especially in areas with irregular demand patterns, such as military logistics. Thus, an advanced model for accurately forecasting this demand was developed in this study. The K-X tank is one of the Republic of Korea Army's third generation main battle tanks. Data about the spare part consumption of 1,053,422 transactional data points stored in a military logistics management system were obtained. Demand forecasting classification models were developed to exploit machine learning, stacked generalization, and time series as baseline methods. Additionally, various stacked generalizations were established in spare part demand forecasting. The results demonstrated that a suitable selection of methods could help enhance the performance of the forecasting models in this domain. [ABSTRACT FROM AUTHOR]

This paper presents a novel method for predicting the traffic speed of the links on large-scale traffic networks. We first analyze how traffic flows in and out of every link through the lowest cost reachable paths. We aggregate the traffic flow conditions of the links on every hop of the inbound and outbound reachable paths to represent the traffic flow dynamics. We compute a new measure called traffic flow centrality (i.e., the Z value) for every link to capture the inherently complex mechanism of the traffic links influencing each other in terms of traffic speed. We combine the features regarding the traffic flow centrality with the external conditions around the links, such as climate and time of day information. We model how these features change over time with recurrent neural networks and infer traffic speed at the subsequent time windows. Our feature representation of the traffic flow for every link remains invariant even when the traffic network changes. Furthermore, we can handle traffic networks with thousands of links. The experiments with the traffic networks in the Seoul metropolitan area in South Korea reveal that our unique ways of embedding the comprehensive spatio-temporal features of links outperform existing solutions. [ABSTRACT FROM AUTHOR]

In terms of electric vehicles (EVs), electric kickboards are crucial elements of smart transportation networks for short-distance travel that is risk-free, economical, and environmentally friendly. Forecasting the daily demand can improve the local service provider's access to information and help them manage their short-term supply more effectively. This study developed the forecasting model using real-time data and weather information from Jeju Island, South Korea. Cluster analysis under the rental pattern of the electric kickboard is a component of the forecasting processes. We cannot achieve noticeable results at first because of the low amount of training data. We require a lot of data to produce a solid prediction result. For the sake of the subsequent experimental procedure, we created synthetic time-series data using a generative adversarial networks (GAN) approach and combined the synthetic data with the original data. The outcomes have shown how the GAN-based synthetic data generation approach has the potential to enhance prediction accuracy. We employ an ensemble model to improve prediction results that cannot be achieved using a single regressor model. It is a weighted combination of several base regression models to one meta-regressor. To anticipate the daily demand in this study, we create an ensemble model by merging three separate base machine learning algorithms, namely CatBoost, Random Forest (RF), and Extreme Gradient Boosting (XGBoost). The effectiveness of the suggested strategies was assessed using some evaluation indicators. The forecasting outcomes demonstrate that mixing synthetic data with original data improves the robustness of daily demand forecasting and outperforms other models by generating more agreeable values for suggested assessment measures. The outcomes further show that applying ensemble techniques can reasonably increase the forecasting model's accuracy for daily electric kickboard demand. [ABSTRACT FROM AUTHOR]

Satellite-based flood monitoring for providing visual information on the targeted areas is crucial in responding to and recovering from river floods. However, such monitoring for practical purposes has been constrained mainly by obtaining and analyzing satellite data, and linking and optimizing the required processes. For these purposes, we present a deep learning-based flood area extraction model for a fully automated flood monitoring system, which is designed to continuously operate on a cloud-based computing platform for regularly extracting flooded area from Sentinel-1 data, and providing visual information on flood situations with better image segmentation accuracy. To develop the new flood area extraction model using deep learning, initial model tests were performed more than 500 times to determine optimal hyperparameters, water ratio, and best band combination. The results of this research showed that at 'waterbody ratio 30%', which yielded higher segmentation accuracies and lower loss, precision, overall accuracy, IOU, recall, and F1 score of 'VV, aspect, topographic wetness index, and buffer input bands' were 0.976, 0.956, 0.894, 0.964, and 0.970, respectively, and averaged inference time was 744.3941 s, which demonstrate improved image segmentation accuracy and reduced processing time. The operation and robustness of the fully automated flood monitoring system were demonstrated by automatically segmenting 12 Sentinel-1 images for the two major flood events in Republic of Korea during 2020 and 2022 in accordance with the hyperparameters, waterbody ratio, and band combinations determined through the intensive tests. Visual inspection of the outputs showed that misclassification of constructed facilities and mountain shadows were extremely reduced. It is anticipated that the fully automated flood monitoring system and the deep leaning-based waterbody extraction model presented in this research could be a valuable reference and benchmark for other countries trying to build a cloud-based flood monitoring system for rapid flood monitoring using deep learning. [ABSTRACT FROM AUTHOR]

The impact of the COVID-19 epidemic on the mental health of elderly individuals is causing considerable worry. We examined a deep neural network (DNN) model to predict the depression of the elderly population during the pandemic period based on social factors related to stress, health status, daily changes, and physical distancing. This study used vast data from the 2020 Community Health Survey of the Republic of Korea, which included 97,230 people over the age of 60. After cleansing the data, the DNN model was trained using 36,258 participants' data and 22 variables. We also integrated the DNN model with a LIME-based explainable model to achieve model prediction explainability. According to the research, the model could reach a prediction accuracy of 89.92%. Furthermore, the F1-score (0.92), precision (93.55%), and recall (97.32%) findings showed the effectiveness of the proposed approach. The COVID-19 pandemic considerably impacts the likelihood of depression in later life in the elderly community. This explainable DNN model can help identify patients to start treatment on them early. [ABSTRACT FROM AUTHOR]

Despite the development of advanced technology, marine accidents have not decreased. To prevent marine accidents, it is necessary to predict accidents in advance. With the recent development of artificial intelligence (AI), AI technologies such as deep learning have been applied to create and analyze predictive models in various fields. The purpose of this study is to develop a model for predicting the frequency of marine accidents using a long-short term memory (LSTM) network. In this study, a prediction model was developed using marine accidents from 1981 to 2019, and the proposed model was evaluated by predicting the accidents in 2020. As a result, we found that marine accidents mainly occurred during the third officer's duty time, representing that the accidents are highly related to the navigator's experience. In addition, the proposed LSTM model performed reliably to predict the frequency of marine accidents with a small mean absolute percentage error (best MAPE: 0.059) that outperformed a traditional statistical method (i.e, ARIMA). This study could help us build LSTM structures for marine accident prediction and could be used as primary data to prevent the accidents by predicting the number of marine accidents by the navigator's watch duty time. [ABSTRACT FROM AUTHOR]

Urban parameters, such as building density and the building coverage ratio (BCR), play a crucial role in urban analysis and measurement. Although several approaches have been proposed for BCR estimations, a quick and effective tool is still required due to the limitations of statistical-based and manual mapping methods. Since a building footprint is crucial for the BCR calculation, we hypothesize that Deep Learning (DL) models can aid in the BCR computation, due to their proven automatic building footprint extraction capability. Thus, this study applies the DL framework in the ArcGIS software to the BCR calculation task and evaluates its efficiency for a new industrial district in South Korea. Although the accuracy achieved was limited due to poor-quality input data and issues with the training process, the result indicated that the DL-based approach is applicable for BCR measuring, which is a step toward suggesting an implication of this method. Overall, the potential utility of this proposed approach for the BCR measurement promises to be considerable. [ABSTRACT FROM AUTHOR]

South Korea currently lacks a real-time monitoring and anomaly detection system for detecting continuous tap water quality changes from the water source to faucet and pre-diagnosing hazards that threaten tap water safety. In this study, we constructed an accurate water quality prediction model that could comprehensively cover all water treatment facilities supplying tap water nationwide and verified the model using an integrated approach. To address the uncertainty of continuously changing water quality, we collected five years (2017–2021) of hourly water quality data from 33 large water purification plants and applied various deep learning techniques to construct an optimal prediction model. We repeated water quality prediction and evaluation over the following 24 h through a time series cross-validation of an untrained dataset of the previous five months. The optimized deep learning model achieved average and maximum prediction accuracy of 98.78 and 99.98%, respectively, and showed excellent performance in terms of the root mean squared error (0.0006), mean absolute error (0.0003), and Nash–Sutcliffe efficiency (0.9894). Thus, deep learning technology greatly improved the accuracy and efficiency of water quality prediction. The proposed model could provide prompt and accurate water quality information for large-scale water supply facilities nationwide and improve public health through the early diagnosis of water quality anomalies. [ABSTRACT FROM AUTHOR]

Remote sensing image retrieval (RSIR) is the process of searching for identical areas by investigating the similarities between a query image and the database images. RSIR is a challenging task owing to the time difference, viewpoint, and coverage area depending on the shooting circumstance, resulting in variations in the image contents. In this paper, we propose a novel method based on a coarse-to-fine strategy, which makes a deep network more robust to the variations in remote sensing images. Moreover, we propose a new triangular loss function to consider the whole relation within the tuple. This loss function improves the retrieval performance and demonstrates better performance in terms of learning the detailed information in complex remote sensing images. To verify our methods, we experimented with the Google Earth South Korea dataset, which contains 40,000 images, using the evaluation metric Recall@n. In all experiments, we obtained better performance results than those of the existing retrieval training methods. Our source code and Google Earth South Korea dataset are available online. [ABSTRACT FROM AUTHOR]

Load power forecast is one of most important tasks in power systems operation and maintenance. Enhancing its accuracy can be helpful to power systems scheduling. This paper presents how to use partial real-time temperature information in forecasting load power, which is usually done using past load power and temperature data. The partial real-time temperature information means temperature information for only part of the entire prediction time interval. To this end, a long short-term memory (LSTM) network is trained using past temperature and load power data in order to forecast load power, where forecasted load power depends on the temperature prediction implicitly. Then, in order to deal with the case where nontrivial temperature prediction errors happen, a multi-layer perceptron (MLP) network is trained using the past data describing the relation between temperature variation and load power variation. Then, the temperature is measured at the beginning of the prediction time-interval and compensated load forecast is computed by adding the output of the LSTM and that of the MLP whose input is the temperature prediction error. It is shown that the proposed compensation using the real-time temperature information indeed improves performance of load power forecast. This improved load forecast is used to predict system marginal price (SMP). The proposed method is validated using the real temperature and load power data of South Korea. [ABSTRACT FROM AUTHOR]