Search

In recent years, active research has been conducted using deep learning to evaluate damage to aging bridges. However, this method is inappropriate for practical use because its performance deteriorates owing to numerous classifications, and it does not use photos of actual sites. To this end, this study used image data from an actual bridge management system as training data and employed a combined learning model for each member among various instance segmentation models, including YOLO, Mask R-CNN, and BlendMask. Meanwhile, techniques such as hyperparameter tuning are widely used to improve the accuracy of deep learning, and this study aimed to improve the accuracy of the existing model through this. The hyperparameters optimized in this study are DEPTH, learning rate (LR), and iterations (ITER) of the neural network. This technique can improve the accuracy by tuning only the hyperparameters while using the existing model for bridge damage identification as it is. As a result of the experiment, when DEPTH, LR, and ITER were set to the optimal values, mAP was improved by approximately 2.9% compared to the existing model.

Currently, damage in aging bridges is assessed visually, leading to significant personnel, time, and cost expenditures. Moreover, the results depend on the subjective judgment of the inspector. Machine-learning-based approaches, such as deep learning, can solve these problems. In particular, instance-segmentation models have been used to identify different types of bridge damage. However, the value of deep-learning-based damage identification may be reduced by insufficient training data, class imbalance, and model-reliability issues. To overcome these limitations, this study utilized photographic data from real bridge-management systems for the inspection and assessment of bridges as the training dataset. Six types of damage were considered. Moreover, the performances of three representative deep learning models—Mask R-CNN, BlendMask, and SWIN—were compared in terms of loss–function values. SWIN showed the best performance, achieving a loss value of 0.000005 after 269,939 training iterations. This shows that bridge-damage-identification performance can be maximized by setting an appropriate learning rate and using a deep learning model with a minimal loss value.

Cyber-security systems collect information from multiple security sensors to detect network intrusions and their models. As attacks become more complex and security systems diversify, the data used by intrusion-detection systems becomes more dimensional and large-scale. Intrusion detection based on intelligent anomaly detection detects attacks based on machine-learning classification models, soft computing, and rule sets. Feature-selection methods are used for efficient intrusion detection and solving high-dimensional problems. Optimized feature selection can maximize the detection model performance; thus, a fitness function design is required. We proposed an optimization algorithm-based feature-selection algorithm to improve anomaly-detection performance. We used a genetic algorithm and proposed an advanced fitness function that finds the most relevant feature set, increasing the detection rate, reducing the error rate, and enhancing analysis speed. An improved fitness function for the selection of optimized features is proposed; this function can address overfitting by solving the problem of anomaly-detection performance from imbalanced security datasets. The proposed algorithm outperformed other feature-selection algorithms. It outperformed the PCA and wrapper-DR methods, with 0.99564 at 10%, 0.996455 at 15%, and 0.996679 at 20%. It performed higher than wrapper-DR by 0.95% and PCA by 3.76%, showing higher differences in performance than in detection rates.

The quality of road pavements is highly impacted by environmental variables, such as temperature, humidity, and weather; and construction-related variables, such as material quality and time. In this paper, an advanced data collection and analysis system based on big data/cloud was proposed for the use of IoT location-based smart platforms for pavement quality big data at road pavement sites. For the big data platform, a relational database management system (RDBMS) for a general alphanumeric data-based infrastructure for IoT-based systems was designed based on distributed/parallel processing to enable rapid and big data collection and analysis. The structure was established based on a NoSQL-based database to enable real-time high-speed collection and analysis, and the big data platform was developed as a data collection and visualization infrastructure. When the big data system was studied using data analysis methods, the proposed system demonstrated improvements in data collection performance and analysis speed, indicating that analysis results could be derived in real time. Specifically, the data collection processing (create) speed of the NoSQL-based system (0.405 ms) was significantly higher than that of the compared existing system (21.146 ms). Real-time processing capacity was also verified based on quality big data generated on actual road pavements, and the proposed system was proven suitable for the real-time monitoring (the data collection processing) of road pavement quality big data.

More bridges today require maintenance with age, owing to increasing structural loads from traffic and natural disasters. Routine inspection for damages, including in the aftermath of special events, is conducted by experts. To address the limitations of human inspection, deep-learning-based analysis of bridge damage is being actively conducted. However, such models exhibit deteriorated performance in classifying multiple classes. Most existing algorithms do not use in situ images. Hence, the results of the model training do not accurately reflect the actual damage. This study utilizes an extant method and proposes a new model of combination training by bridge member. By integrating the two approaches, we propose a bridge damaged-object-detection deep-combination framework (BDODC-F). To ensure variety in the type of damaged objects and enhanced model performance, a deep-learning-based super-resolution module is employed. For performance improvement and optimization, a deep-learning combination model based on individual training by bridge member is proposed. The BDODC-F improved the mean average precision by 191.6% and 112.21% in the combination model. We expect the framework to aid engineers in the automated detection and identification of bridge damage.

The growing dependence on digital systems has heightened the risks posed by cybersecurity threats. This paper proposes a new method for detecting malicious webpages among several adversary activities. As shown in previous studies, malicious URL detection performance is significantly affected by the learning dataset features. The overall performance of different machine learning models varies depending on the data features, and using a particular model alone is not always desirable in any given environment. To address these limitations, we propose an ensemble approach using different machine learning models. Our proposed method outperforms the existing single model by 6%, allowing for the detection of an additional 141 malicious URLs. In this study, repetitive tasks are automated, improving the performance of different machine learning models. In addition, the proposed framework builds an advanced feature set based on URL and web content and includes the most optimized detection model structure. The proposed technology can contribute to define an advanced feature set based on URL and web content and includes the most optimized detection model structure and research on automated technology for the detection of malicious websites, such as phishing websites and malicious code distribution.

With the rapid development of global trade and maritime transportation, a comprehensive emission inventory is necessary to control air pollution from ships as well as to comply with international environmental standards. The purpose of this study is to calculate the volume of emissions from ships in Yeosu and Gwangyang ports in Korea based on a bottom-up approach and supported by the data from the automatic identification system (AIS) and Korean port management information system (Port-MIS). Specifically, a real set of data on specifications of all ships operating at the port in 2019 was collected and an analysis of ship movement using AIS data was conducted by the authors in this study to divide the route into cruising and maneuvering phases as well as to estimate engine’s load factor. Finally, results show that the total amount of air emissions from ships at the port area was 558 thousand tons, which is 96% from CO2 and 66% from hoteling mode. Additionally, container and tanker ships together emitted the highest volume of air pollution accounting for a share of 73%.

As network technology has advanced, and as larger and larger quantities of data are being collected, networks are becoming increasingly complex. Various vulnerabilities are being identified in such networks, and related attacks are continuously occurring. To solve these problems and improve the overall quality of network security, a network risk scoring technique using attack graphs and vulnerability information must be used. This technology calculates the degree of risk by collecting information and related vulnerabilities in the nodes and the edges existing in the network-based attack graph, and then determining the degree of risk in a specific network location or the degree of risk occurring when a specific route is passed within the network. However, in most previous research, the risk of the entire route has been calculated and evaluated based on node information, rather than edge information. Since these methods do not include correlations between nodes, it is relatively difficult to evaluate the risk. Therefore, in this paper, we propose a vulnerability Correlation and Attack Graph-based node-edge Scoring System (VCAG-SS) that can accurately measure the risk of a specific route. The proposed method uses the Common Vulnerability Scoring System (CVSS) along with node and edge information. Performing the previously proposed arithmetic evaluation of confidentiality, integrity, and availability (CIA) and analyzing the correlation of vulnerabilities between each node make it possible to calculate the attack priority. In the experiment, the risk scores of nodes and edges and the risk of each attack route were calculated. Moreover, the most threatening attack route was found by comparing the attack route risk. This confirmed that the proposed method calculated the risk of the network attack route and was able to effectively select the network route by providing the network route priority according to the risk score.

Rancidity data can be used in various fields such as the quality analysis of food and raw materials used for construction. The rancidity of raw materials used in road pavement asphalt is currently only at the level determined by the temperature or visual factors. Although construction workers are managed individually and subjectively, such as by visual methods, they cannot be managed in practice. In this paper, we propose a system combining a rancidity sensor with an Internet of Things (IoT) communication module that collects and predicts rancidity measurements in real time at a site. The values measured by the sensor are periodically transferred to the Cloud through the IoT communication module, the validity of the data set is established, and the systematic management of the data is performed using machine-learning-based data analysis techniques. The results of an experiment showed a high classification prediction accuracy of 91.3% and a short-term pattern prediction accuracy of 96.6% (weighted scaling), confirming its excellent potential for raw material quality management. The results of this paper will be applied as a road pavement quality management system. [ABSTRACT FROM AUTHOR]

This study was conducted to develop a predictive model to estimate the growth of Listeria monocytogenes on fresh pork during storage at constant temperatures (5, 10, 15, 20, 25, 30, and 35°C). The Baranyi model was fitted to growth data (log CFU per gram) to calculate the specific growth rate (SGR) and lag time (LT) with a high coefficient of determination (R² > 0.98). As expected, SGR increased with a decline in LT with rising temperatures in all samples. Secondary models were then developed to describe the variation of SGR and LT as a function of temperature. Subsequently, the developed models were validated with additional independent growth data collected at 7, 17, 27, and 37°C and from published reports using proportion of relative errors and proportion of standard error of prediction. The proportion of relative errors of the SGR and LT models developed herein were 0.79 and 0.18, respectively. In addition, the standard error of prediction values of the SGR and LT of L. monocytogenes ranged from 25.7 to 33.1% and from 44.92 to 58.44%, respectively. These results suggest that the model developed in this study was capable of predicting the growth of L. monocytogenes under various isothermal conditions. [ABSTRACT FROM AUTHOR]

Big Data means a very large amount of data and includes a range of methodologies such as big data collection, processing, storage, management, and analysis. Since Big Data Text Mining extracts a lot of features and data, clustering and classification can result in high computational complexity and the low reliability of the analysis results. In particular, a TDM (Term Document Matrix) obtained through text mining represents term-document features but features a sparse matrix. In this paper, the study focuses on selecting a set of optimized features from the corpus. A Genetic Algorithm (GA) is used to extract terms (features) as desired according to term importance calculated by the equation found. The study revolves around feature selection method to lower computational complexity and to increase analytical performance. We designed a new genetic algorithm to extract features in text mining. TF-IDF is used to reflect document-term relationships in feature extraction. Through the repetitive process, features are selected as many as the predetermined number. We have conducted clustering experiments on a set of spam-mail documents to verify and to improve feature selection performance. And we found that the proposal FSGA algorithm shown better performance of Text Clustering and Classification than using all of features. [ABSTRACT FROM AUTHOR]