Search

Software vulnerabilities in web applications are one of the sensitive points in data and application security. Although closing a vulnerability after it is detected in web applications seems to be a solution, detecting vulnerabilities in the source code before the vulnerability is detected effectively prevents malicious attacks. In this paper, we present an improved and automated Bidirectional Encoder Representations from Transformers (BERT)-based approach to detect vulnerabilities in web applications developed in C-Sharp. For the training and testing of the proposed VULREM (Vulnerability Remzi) model, a dataset of eight different CVE (Common Vulnerabilities and Exposures)-numbered critical vulnerabilities was created from the source code of six different applications specific to the study. In the VULREM model, fine-tuning was performed within the BERT model to obtain maximum accuracy from the dataset. To obtain the optimum performance according to the number of source-code lines, six different input lengths were tested with different batch sizes. Classification metrics were used for the testing and performance evaluation of the model, and an average F1-score of 99% was obtained for the best sequence length according to eight different vulnerability classifications. In line with the findings obtained, this will play an important role in both vulnerability detection in web applications of the C-Sharp language and in detecting and correcting critical vulnerabilities in the developmental processes of web applications, with an accuracy of 99%. [ABSTRACT FROM AUTHOR]

Due to the random drift of the target aiming point of the detection system on large undersea vehicles, high-speed attack-type unmanned undersea vehicles(UUVs) often cross the front and rear sides of large undersea vehicles vertically in engineering applications, resulting in target miss. To address the above issues, this article proposed a new vertical hit guidance method that utilized multi-cycle data from the detection system to achieve vertical hits in the middle of a large undersea vehicle. In addition, statistical simulation was conducted. The simulation results show that this method has a relatively low sensitivity to guidance parameters and strong adaptability to the relative position of UUVs and the target vehicle. It can meet the requirements of vertical hit engineering under the existing guidance parameter accuracy requirements. When the hit location is considered, the middle of the large undersea vehicle can be hit vertically, and there is a significant increase in the probability of vertical hit. The method proposed in the article is reasonable and feasible and can improve the damage effect on large undersea vehicles.

Assume that a graph G$$ G $$ models a detection system for a facility with a possible "intruder," or a multiprocessor network with a possible malfunctioning processor. We consider the problem of placing detectors at a subset of vertices in G$$ G $$ to determine the location of an intruder if there is any. Many types of detection systems have been defined for different sensor capabilities; in particular, we focus on identifying codes, where each detector can determine whether there is an intruder within its closed neighborhood. In this research we explore a fault‐tolerant variant of identifying codes applicable to real‐world systems. Specifically, error‐detecting identifying codes permit a false‐negative transmission from any single detector. We investigate minimum‐sized error‐detecting identifying codes in several classes of graphs, including cubic graphs and infinite grids, and show that the problem of determining said minimum size in arbitrary graphs is NP‐complete. [ABSTRACT FROM AUTHOR]

Most fire spread is caused by the absence of suppression means at the beginning of the fire. This results in the missed golden time. There are various factors that cause initial fires, such as electrical outlets, general distribution circuits, and oil–vapor–gas cluster spaces. In most cases, these places are out of reach of human hands or they lose the initial suppression time when a fire occurs, causing the spread of fire. This study implements an intelligent fire suppression device that connects sensor IoT based on microcapsules to secure initial fire suppression and golden time in the event of a fire in blind spots that cannot be seen by humans or at a time when it is difficult to recognize a fire. The microcapsule is a micro-collection unit that collects Novec 1230 gas generated in the semiconductor production process. The microcapsule is molded into a form with a fire suppression function and, when a fire occurs, the molded body explodes and absorbs ambient oxygen to suppress the fire. The complex-sensor IoT executes smoke and heat detection generated when a fire is suppressed within 10 s, which ensures the reliability of the detector by notifying of the fire and detecting the ignition point through communication linkages such as Ieee 485 and WiFi or LoRa. [ABSTRACT FROM AUTHOR]

Copyright of Experimental Technology & Management is the property of Experimental Technology & Management Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

During the sintering process in iron production, wheel swing is a sign of sintering machine trolley axle faults, which may lead to the wheel falling off and affect the production operation of the sintering machine system in serious cases. To solve this problem, this paper proposes a fault detection and localization method based on the You Only Look Once version 9 (YOLOv9) object detection algorithm and frame difference method for detecting sintering machine trolley wheel swing. The wheel images transmitted from the camera were sent to a trolley wheel and side panel number detection model that was trained on YOLOv9 for recognition. The wheel recognition boxes of the previous and subsequent frames were fused into the wheel region of interest. In the wheel region of interest, the difference operation was carried out. The result of the difference operation was compared with the preset threshold to determine whether the trolley wheel swings. When a wheel swing fault occurs, the image of the side plate at the time of the fault is collected, and the number on the side plate is identified so as to accurately locate the faulty trolley and to assist the field personnel in troubleshooting the fault. The experimental results show that this method can detect wheel swing faults in the industrial field, and the detection accuracy of wheel swing faults was 93.33%. The trolley side plate numbers' average precision was 99.2% in fault localization. Utilizing the aforementioned method to construct a system for detecting wheel swing can provide technical support for fault detection of the trolley axle on the sintering machine. [ABSTRACT FROM AUTHOR]

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

In order to identify and combat misinformation on social networking sites, the present study analyses the use of automated learning algorithms. With the appearance of the internet and its extensive use, social media structures have turn out to be massive venues for statistics dissemination and influence. But, these platforms have additionally been exploited for spreading propaganda and false data, concentrated on people, corporations, and political entities. Our study specializes in analysing diverse machine learning classifiers, which include Support Vector Machines (SVM), Random forest, and deep learning strategies like BERT and Roberta, to differentiate among propaganda and non-propaganda content. We make use of datasets collected from online information resources and Twitter, leveraging the Twitter API for information extraction. The paper presents a complete evaluate of cutting-edge methodologies, highlights the demanding situations faced in propaganda detection, and discusses the effectiveness of different machine learning models based on their accuracy and applicability in real-world eventualities. Our findings indicate that machine learning, especially advanced models like RoBERTa, can appreciably useful resource in figuring out and mitigating the spread of propaganda and extremist content on social networks. [ABSTRACT FROM AUTHOR]

With the growing integration of IoT devices in critical infrastructure, cybersecurity threats such as Distributed Denial of Service (DDoS) attacks on Energy Hubs (EH) have become a significant concern. This study aims to address these challenges by evaluating the effectiveness of various supervised machine learning (ML) algorithms in predicting DDoS attacks targeting EH systems through IoT devices. Using the CICDDOS2019 and KDD-CUP datasets, a comprehensive analysis was conducted on several classifiers, including Decision Tree (DT), Gradient Boosting, Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Random Forest. The results highlight Gradient Boosting as the most effective model, particularly for the CICDDOS2019 dataset, demonstrating superior accuracy and predictive capability. Additionally, hybrid models combining Gradient Boosting with SVM or DT showed strong performance, though with varying precision and recall. This study provides valuable insights into the selection and tailoring of ML models for specific security challenges, emphasizing the need for ongoing research to enhance the resilience of EH systems and IoT devices against evolving DDoS threats.

Forest fire is a significant issue, especially for tropical countries like Indonesia. One of the impacts of forest fires is environmental pollution and damage, such as damage to flora and fauna, water, and soil. Fire detection technology is crucial as a preventive measure before the spread or expansion of fire points. Several forest fire detection systems have been developed by various research studies, with detection targets varying. Objects in the form of images are usually detected using the RGB color filtering method, but this method still results in false detections in image processing. Therefore, a classification model is built to detect fire and smoke in images using the Convolutional Neural Network (ConvNet) algorithm. In the development of the ConvNet model, a comparison of models is also conducted to assess the influence of Hyperparameter Tuning and Callbacks in optimizing the model's classification performance. The research results indicate that out of the six comparison scenarios created, the best model is obtained with 90% training data and 10% testing data, which is also optimized with Hyperparameter Tuning and Callbacks, with a Validation Accuracy of 98.18% and Validation Loss of 4.97%. This model is then implemented in the interface system.

Group decision-making involves selecting among limited options. Experts share their perspectives in a comparative debate but then evaluate alternatives using preference relations, which can result in inconsistencies between their expressions and assessments. To address this, a method is proposed that automates the generation of these relationships from the debate comments, classifying them into positive and negative using sentiment analysis, namely with the Large Language Model. A new operator is introduced that weights these comments to calculate preference relations. Furthermore, modification of the relationships is allowed if the experts so wish. Moreover, another operator is incorporated that adjusts the weight of each expert according to his or her active participation in the discussion, assigning more weight to those who contribute more comments. Finally, this innovative method promotes coherence and equal participation in group decision-making by employing an innovative sentiment analysis detection system. [ABSTRACT FROM AUTHOR]

Acute respiratory infections (ARI) are infectious diseases that affect both children and adults, particularly in the context of climate change. Bacteria are one of the causes of ARI. According to the government, the discovery of the bacteria that cause ARI is an indicator of successful management of infectious diseases. The current obstacle is the limited number of medical analysts, which results in longer microscopic examination times and requires a high level of objectivity. Therefore, a system for the early detection of ARI-causing bacteria was developed using digital image processing techniques, specifically channel area thresholding as one of the segmentation methods. This research employs four shape features for bacterial classification: the number of bacterial colonies, area, perimeter, and shape. The Naïve Bayes intelligent system method is used for the classification process. The system had an accuracy rate of 86.84% in the classification of four types of bacteria: S. aureus, S. pneumoniae, C. diphteriae and M. tuberculosis. [ABSTRACT FROM AUTHOR]

Stored grain is a complex ecosystem in which intricate multi-field interactions exist among abiotic and biotic factors, as well as the surrounding environment. Clearly understanding these interactions is crucial for ensuring grain storage security. This study developed a graphical detection system based on low-field nuclear magnetic resonance (LF-NMR) technology, which consists of an LF-NMR imaging analyzer, a small grain container, and dedicated software. This system can simultaneously detect the temperature, moisture, and humidity of a grain sample stored in the grain container and visually present the cloud maps of these fields through the dedicated software. To verify the system's performance, two laboratory storage experiments with paddy rice samples were conducted for 15 d. The results indicated that the measured cloud maps could accurately depict the variations in the temperature, moisture, and humidity fields within the stored paddy rice samples during the storage period. The areas with potential risks of fungal growth, grain sprouting, and moisture condensation due to the multi-field interactions could also be identified through the cloud maps, which demonstrated the credible performance of the system. This system could provide a new technical means to uncover the complex coupling relationships within grain storage ecosystems. • The first use of LF-NMR to investigate the multi-field interactions in stored grain. • A novel graphical detection system based on LF-NMR was developed. • The system was able to detect three key physical fields and generate cloud maps. • Two storage experiments were conducted to validate the system's performance. [ABSTRACT FROM AUTHOR]

3D printed ceramic parts often have defects due to their inherent brittleness. These defects are of various scale sizes. Detecting these defects, especially with multiple sizes, is a challenging task in the field of detection. Furthermore, there is a lack of surface defect real-time detection systems suitable for industrial applications in this field. Based on this, this paper makes two contributions: the design and development of a real-time defect detection system, and the proposal of a multiscale defect detection method for 3D printed ceramic surfaces within this system. Firstly, the paper establishes the overall structure of the real-time detection system for surface defects on 3D printed ceramic components, and describes the hardware and software components of the system. On this basis, a dataset of ceramic surface defect images is collected and constructed. Then, experimental analyses point out the shortcomings of the You Only Look Once version-5 (YOLOv5) model for multiscale defect detection. To address the shortcomings, the YOLOv5 model is optimized from three aspects, resulting in the Deep separable convolution + residual network-SKNetwork-Efficient Channel Attention Network-YOLOv5 (DepRes-SK-ECA-YOLOv5) model for multiscale defect detection. This model improves the ability to extract and fuse features of defects at different scales. The experimental results show that the DepRes-SK-ECA-YOLOv5 model can achieve 93.5 %, 91.6 %, 94.3 %, and 0.198 s for Precision, Recall, mAP, and Speed for the test set, respectively. Finally, the paper designs the workflow for the system software. The system hardware and system software are integrated to form a real-time detection system. The performance of the detection system is verified through experiments. [ABSTRACT FROM AUTHOR]

A detection system, modeled in a graph, is composed of "detectors" positioned at a subset of vertices in order to uniquely locate an "intruder" at any vertex. Identifying codes use detectors that can sense the presence or absence of an intruder within distance one. We introduce a fault-tolerant identifying code called a redundant identifying code, which allows at most one detector to go offline or be removed without disrupting the detection system. In real-world applications, this would be a necessary feature, as it would allow for maintenance on individual components without disabling the entire system. Specifically, we prove that the problem of determining the lowest cardinality of a redundant identifying code for an arbitrary graph is NP-hard, and we determine the bounds on the lowest cardinality for special classes of graphs, including trees, ladders, cylinders, and cubic graphs. [ABSTRACT FROM AUTHOR]

Software vulnerabilities in web applications are one of the sensitive points in data and application security. Although closing a vulnerability after it is detected in web applications seems to be a solution, detecting vulnerabilities in the source code before the vulnerability is detected effectively prevents malicious attacks. In this paper, we present an improved and automated Bidirectional Encoder Representations from Transformers (BERT)-based approach to detect vulnerabilities in web applications developed in C-Sharp. For the training and testing of the proposed VULREM (Vulnerability Remzi) model, a dataset of eight different CVE (Common Vulnerabilities and Exposures)-numbered critical vulnerabilities was created from the source code of six different applications specific to the study. In the VULREM model, fine-tuning was performed within the BERT model to obtain maximum accuracy from the dataset. To obtain the optimum performance according to the number of source-code lines, six different input lengths were tested with different batch sizes. Classification metrics were used for the testing and performance evaluation of the model, and an average F1-score of 99% was obtained for the best sequence length according to eight different vulnerability classifications. In line with the findings obtained, this will play an important role in both vulnerability detection in web applications of the C-Sharp language and in detecting and correcting critical vulnerabilities in the developmental processes of web applications, with an accuracy of 99%.

Despite cutting-edge approaches for detecting spam reviews, there is still a lack of precision to classify new reviews since spammer strategy/behavior rapidly evolves and goes back over time. Most existing studies are based on static models that cannot detect spammers with new spam behaviors. This study tackles this challenge for the first time and proposes a Dynamic Spam Detection System (DSDS) to improve the detection model over time dynamically. The DSDS system integrates the neural network (NN) with reinforcement learning (RL) to identify spammer assaults in offline and online modes. In offline mode, the RL aims to identify the best NN architecture for the offline/static consumer review dataset. In the online mode, the DSDS system has the merit of handling the limited dataset problem by automatically adding new reviews to the offline dataset. Moreover, a novel feature selection-based algorithm is proposed to explore new spammer behaviors from the new reviews. The experiments that were conducted rigorously using well-known datasets demonstrated the ability of the system to detect new spam behaviors as well as to effectively classify online reviews by achieving a high accuracy rate and a low false-positive rate of 94.23%, and 0.0026%, respectively, for the YelpChi dataset, and of 97% and 0.0021, respectively, for the Amazon dataset. Moreover, a comparison with the state-of-the-art approaches on the same datasets proves the contribution of the proposed system. [ABSTRACT FROM AUTHOR]

Copyright of Computer Measurement & Control is the property of Magazine Agency of Computer Measurement & Control and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Farmers and producers may suffer significant financial losses due to plant diseases. Effective control of plant diseases requires early identification and categorization of the illnesses. However, traditional plant disease detection techniques are frequently labor- and time-intensive. Plant disease detection has found a possible collaborator in deep learning and machine learning. Consequently, a novel deep-learning technique is proposed for classifying and diagnosing plant illnesses. It uses a transfer learning methodology built on a convolutional neural network architecture consisting of 5 convolutional, 2 fully- connected, and 3 max pooling layers. The algorithm has a 98% accuracy rate when tested on a publicly accessible dataset of images of plant diseases. The CNN model was trained with the PlantVillage dataset, which is available on Kaggle. From the results and a comparison with previous works, it is evident that the chosen approach performed well in classifying plant diseases with expected accuracy. [ABSTRACT FROM AUTHOR]

Traditional fire assay analysis has always used manual work methods, with high labor intensity, low efficiency, and high occupational health and safety risks. The developed automated fire assay detection system consists of an automated batching system, automated mixing and filling system, automated melting system, automated cupellation system, and automated gold distribution system, corresponding to 5 analysis steps of batching, mixing, melting, cupellation, and gold distribution. The system realizes the automated detection mode of fire assay, with accurate and reliable detection results, consistent with the manual detection results, and a unified and stable operation process, free from subjective factors of personnel. After being put into production and application, the system greatly improves the detection efficiency, increasing it by more than 167 percentage points compared with manual mode, reducing labor intensity, and ensuring the occupational health and safety of personnel. The automated fire assay detection system has a high value for promotion and application. [ABSTRACT FROM AUTHOR]

In order to cope with the increasing incidence of network attack events, a network attack detection algorithm based on semi-supervised learning was designed through self-training based on adaptive enhancement algorithms. A network attack detection system was designed based on this algorithm, which mainly included data acquisition, processing and well as detection units. The experimental results show that on the KDDTest dataset, the proposed algorithm outperforms the semi-supervised STBoot algorithm in terms of accuracy, precision, and recall. Which meets the requirements of the design accuracy for the network attack detection system. [ABSTRACT FROM AUTHOR]

One of the functional vectors of the Cybersecurity Strategy of Ukraine is the development and implementation of protection systems for various information platforms in society's infrastructure, particularly focusing on creating safe technologies to detect deepfake-modifications of biometric images, based on neural networks in cyberspace. This paper presents the security principles of neural network information technologies (IT) within the context of deepfake-modifications. It delineates a basic approach for safely detecting deepfake-modifications in biometric images and outlines a security methodology for multi-level neural network IT systems, organized according to the "object – threat – protection" concept. The basic approach integrates information neural network technology with decision support IT, structured by a modular architecture for detecting deepfake modifications. This architecture operates across the stages of "pre-processing of feature data – classifier training." The core of the IT security methodology emphasizes the integrity of neural network systems for detecting deepfakemodifications in biometric images, coupled with data analysis systems that execute the information process of "dividing video files into frames – detecting and processing features – evaluating the accuracy of image classifiers. The security methodology for multi-level neural network IT relies on systemic and synergistic approaches to construct a comprehensive IT security system. This system accounts for the possibility of emergent threats and incorporates cutting-edge countermeasure technologies at both hardware and software levels. The proposed comprehensive security system for detecting deepfake-modifications in biometric images encompasses hardware and software tools across several segments: automated classifier accuracy assessment, real-time deepfake-modification detection, sequential image processing, and classification accuracy evaluation utilizing cloud computing. [ABSTRACT FROM AUTHOR]

This study involved the preparation of an all-solid-state ion-selective electrode (ASS-ISE) with copper and a poly(3,4-ethylenedioxythiophene) and polystyrene sulfonate (PEDOT/PSS) conversion layer through electrode deposition. The morphology of the PEDOT/PSS film was characterized, and the performance of the copper ion-selective film was optimized. Additionally, a microfluidic chip for the ASS-ISE with copper was designed and prepared. An integrated microfluidic chip test system with an ASS-ISE was developed using a self-constructed potential detection device. The accuracy of the system was validated through comparison testing with atomic absorption spectrophotometry (AAS). The experimental findings indicate that the relative standard deviation (RSD) of the integrated ASS-ISE with the copper microfluidic chip test system is 4.54%, as compared to the industry standard method. This value complies with the stipulated requirement of an RSD ≤ 5% in DL/T 955-2016. [ABSTRACT FROM AUTHOR]

A Cluster-Based Wireless Sensor Network (CBWSN) is a system designed to remotely control and monitor specific events or phenomena in areas such as smart grids, intelligent healthcare, circular economies in smart cities, and underwater surveillance. The wide range of applications of technology in almost every field of human activity exposes it to various security threats from cybercriminals. One of the pressing concerns that requires immediate attention is the risk of security breaches, such as intrusions in wireless sensor network traffic. Poor detection of denial-of-service (DoS) attacks, such as Grayhole, Blackhole, Flooding, and Scheduling attacks, can deplete the energy of sensor nodes. This can cause certain sensor nodes to fail, leading to a degradation in network coverage or lifetime. The detection of such attacks has resulted in significant computational complexity in the related works. As new threats arise, security attacks get more sophisticated, focusing on the target system's vulnerabilities. This paper proposed the development of Cluster-Based Wireless Sensor Network and Variable Selection Ensemble Machine Learning Algorithms (CBWSN_VSEMLA) as a security threats detection system framework for DoS attack detection. The CBWSN model is designed using a Fuzzy C-Means (FCM) clustering technique, whereas VSEMLA is a detection system comprised of Principal Component Analysis (PCA) for feature selection and various ensemble machine learning algorithms (Bagging, LogitBoost, and RandomForest) for the detection of grayhole attacks, blackhole attacks, flooding attacks, and scheduling attacks. The experimental results of the model performance and complexity comparison for DoS attack evaluation using the WSN-DS dataset show that the PCA_RandomForest IDS model outperforms with 99.999 % accuracy, followed by the PCA_Bagging IDS model with 99.78 % accuracy and the PCA_LogitBoost model with 98.88 % accuracy. However, the PCA_RandomForest model has a high computational complexity, taking 231.64 s to train, followed by the PCA_LogitBoost model, which takes 57.44 s to train, and the PCA_Bagging model, which takes 0.91 s to train to be the best in terms of model computational complexity. Thus, the models surpassed all baseline models in terms of model detection accuracy on flooding, scheduling, grayhole, and blackhole attacks.

Never happened before in human history the spreading of fake news; now, the development of the Worldwide Web and the adoption of social media have given a pathway for people to spread misinformation to the world. Everyone is using the Internet, creating and sharing content on social media, but not all the information is valid, and no one is verifying the originality of the content. Identifying the content's essence is sometimes complicated for researchers and intelligent researchers. For example, during Covid-19, misinformation spread worldwide about the outbreak, and much false information spread faster than the virus. This misinformation will create a problem for the public and mislead people into taking the proper medicine. This work will help us to improve the prediction rate. Here we investigate the ability of machine learning classifiers and deep learning models: Naive Bayes, Logistic Regression, Support Vector Machine, Decision Tree, Random Forest and K-Nearest Neighbor. Deep learning models include Convolutional Neural Networks and Long Short-Term Memory (LSTM). The various types of machine learning and deep learning models will be trained and tested using the Covid-19 dataset (1,375,592 tweets).

Objective To verificate the performance of enzyme-linked immunosorbent assay (ELISA) in blood screening laboratory. Methods The repeatability, precision, sensitivity, specificity, compliance, detection limit and anti-interference of ELISA items in the laboratory detection system were verified. Results The repeatability was 100%.The intra batch imprecision of each system was less than 10%, and the inter batch imprecision was less than 15%. The sensitivity, specificity and compliance were 100%, with the minimum detection limits of the two reagents at 0.75 NCU/mL and 0.25 NCU/mL respectively, The anti-interference met the requirements of the reagent manual. Conclusion The analysis of the performance verification data of ELISA test items will help continuously improve the performance of detection system and ensure the safety of clinical blood use.

Most fire spread is caused by the absence of suppression means at the beginning of the fire. This results in the missed golden time. There are various factors that cause initial fires, such as electrical outlets, general distribution circuits, and oil–vapor–gas cluster spaces. In most cases, these places are out of reach of human hands or they lose the initial suppression time when a fire occurs, causing the spread of fire. This study implements an intelligent fire suppression device that connects sensor IoT based on microcapsules to secure initial fire suppression and golden time in the event of a fire in blind spots that cannot be seen by humans or at a time when it is difficult to recognize a fire. The microcapsule is a micro-collection unit that collects Novec 1230 gas generated in the semiconductor production process. The microcapsule is molded into a form with a fire suppression function and, when a fire occurs, the molded body explodes and absorbs ambient oxygen to suppress the fire. The complex-sensor IoT executes smoke and heat detection generated when a fire is suppressed within 10 s, which ensures the reliability of the detector by notifying of the fire and detecting the ignition point through communication linkages such as Ieee 485 and WiFi or LoRa.

During the sintering process in iron production, wheel swing is a sign of sintering machine trolley axle faults, which may lead to the wheel falling off and affect the production operation of the sintering machine system in serious cases. To solve this problem, this paper proposes a fault detection and localization method based on the You Only Look Once version 9 (YOLOv9) object detection algorithm and frame difference method for detecting sintering machine trolley wheel swing. The wheel images transmitted from the camera were sent to a trolley wheel and side panel number detection model that was trained on YOLOv9 for recognition. The wheel recognition boxes of the previous and subsequent frames were fused into the wheel region of interest. In the wheel region of interest, the difference operation was carried out. The result of the difference operation was compared with the preset threshold to determine whether the trolley wheel swings. When a wheel swing fault occurs, the image of the side plate at the time of the fault is collected, and the number on the side plate is identified so as to accurately locate the faulty trolley and to assist the field personnel in troubleshooting the fault. The experimental results show that this method can detect wheel swing faults in the industrial field, and the detection accuracy of wheel swing faults was 93.33%. The trolley side plate numbers’ average precision was 99.2% in fault localization. Utilizing the aforementioned method to construct a system for detecting wheel swing can provide technical support for fault detection of the trolley axle on the sintering machine.

For didactic purposes, the aim of this work was to improve a simulation model of surface electromyographic (sEMG) signal by taking into consideration the shortcomings of previously developed models. This model started with the simulation of the single fibre action potential (SFAP), then the model of the single motor unit action potential (MUAP), afterwards the imitation of the train of MUAP and finally the modellig of the resultant sEMG signal which is the sum of the MUAPs trains. SFAP simulation was based on: i) the description of the volume conductor model which is composed of four layers (bone, muscle, fat and skin), ii) the description of the electrodes shapes and sizes as well as spatial filters, iii) and the transmebrane current. The proposed model shows its effectiveness in the possibility of carrying out practical work by simulation on the modelling of SFAP, MUAP, MUAPT and the sEMG signal. The most important result of this model is that signal processing tools can be applied to analyze and interpret real-world phenomena such as the effects of physiological, non physiological and sensing system parameters on the shape of the simulated sEMG signal. [ABSTRACT FROM AUTHOR]

The rapid adoption of the Internet of Things (IoT) across industries has revolutionized daily life by providing essential services and leisure activities. However, the inadequate software protection in IoT devices exposes them to cyberattacks with severe consequences. Intrusion Detection Systems (IDS) are vital in mitigating these risks by detecting abnormal network behavior and monitoring safe network traffic. The security research community has shown particular interest in leveraging Machine Learning (ML) approaches to develop practical IDS applications for general cyber networks and IoT environments. However, most available datasets related to Industrial IoT suffer from imbalanced class distributions. This study proposes a methodology that involves dataset preprocessing, including data cleaning, encoding, and normalization. The class imbalance is addressed by employing the Synthetic Minority Oversampling Technique (SMOTE) and performing feature reduction using correlation analysis. Multiple ML classifiers, including Logistic Regression, multi-layer perceptron, Decision Trees, Random Forest, and XGBoost, are employed to model IoT attacks. The effectiveness and robustness of the proposed method evaluate using the IoTID20 dataset, which represents current imbalanced IoT scenarios. The results highlight that the XGBoost model, integrated with SMOTE, achieves outstanding attack detection accuracy of 0.99 in binary classification, 0.99 in multi-class classification, and 0.81 in multiple sub-classifications. These findings demonstrate our approach’s significant improvements to attack detection in imbalanced IoT datasets, establishing its superiority over existing IDS frameworks. [ABSTRACT FROM AUTHOR]

This paper studied the problem of ice and snow pollutants identification and freezing temperature detection under winter runway operation, and designed an active detection system for runway ice and snow pollutants and freezing temperature. The temperature change trend during the liquid freezing process was researched, and the freezing temperature detection model based on sequence segmented linear fitting and inflection point identification was proposed in combination with active cooling technology. The differences in dielectric properties of runway snow and ice cover are studied, and a multi-frequency detection-based forked-finger planar capacitive sensor was identified for snow and ice pollutants detection considering practical application scenarios. A finite element simulation model of the forked-finger planar capacitive sensor was established, to optimize the structural parameters of the sensor and verify the feasibility of the capacitor sensor. Finally, the detection device and control system were designed and fabricated, and built an experimental platform for system test. The field experimental results showed that the system has good reliability and stability, and the error of freezing temperature detection model is less than 0.3 °C. At the same time, it can identify three types of pollutants, including water, ice and ice–water mixture, with an accuracy of 89%. [ABSTRACT FROM AUTHOR]

In construction engineering safety management, the problem of construction workers' unsafe behavior (CWUB) has always been a focus for researchers as well as practice managers. Currently, most studies focus on the influencing factors and mechanisms of (CWUB), with less attention given to the dissemination process and control effects of CWUB. Therefore, this paper aims to investigate a safety control detection system for the transmission process. The heterogeneous cellular automaton (CA) has advantages in constructing such a system as it can reflect the interactive processes of construction workers from micro to macro, local to global, and consider the heterogeneity of individuals and space, satisfying unequal interaction probabilities between individuals and spatial variations in characteristics. The SEIR model accurately categorizes construction workers and visually represents the changing quantities of different state groups at each stage. It effectively describes the process of CWUB transmission among construction workers. Based on the aforementioned foundation, a safety control and monitoring system was proposed for the implementation stages of the project. Finally, the control detection system is simulated to assess its effectiveness. Simulation results closely align with reality, showing a continuous decrease in susceptible individuals, a peak followed by a rapid decline in latent and infected individuals, and a steady increase in immune individuals. To control CWUB transmission, it is crucial to enhance immunity against unsafe behaviors, reduce the rate of immunity conversion, and shorten the disease cycle caused by such behaviors. This research has practical implications for construction projects. [ABSTRACT FROM AUTHOR]

This research introduces depression as a prevalent mental health condition affecting millions of people worldwide. A sentiment analysis framework is developed for Facebook to detect signs of depression within user posts. The system uses NLP (natural language processing) and machine learning algorithms (CNN) to analyze sentiment and classify posts as positive, neutral, or negative. The framework integrates into Facebook's infrastructure, enhancing accuracy and efficiency. It incorporates user-specific contextual information and performs comparative analyses against existing methods and clinical evaluations. The results show the system effectively identifies posts indicative of depressive sentiments with high accuracy and sensitivity. The sentiment analysis framework can be adapted and implemented in various social media platforms, facilitating proactive mental health interventions, and supporting individuals in need. Integrating the system into digital health solutions can contribute to a more comprehensive approach to mental health care, reaching a wider population and providing timely support. [ABSTRACT FROM AUTHOR]

The article presents the results of experiments on a detection system used for detecting signals from a miniature, low-energy micro-electro-mechanical system (MEMS) X-ray source. The authors propose to use a detection based on luminescence phenomena occurring in luminophore and scintillators to record the visual signal on a CMOS/CCD detector. The main part of the article is a review of various materials of scintillators and luminophores which would be adequate to convert low-energy X-ray radiation (E < 25 keV - it is a range not typical for conventional X-ray systems) to visible light. Measurements obtained for different energies, exposure times, and different targets have been presented and analysed. [ABSTRACT FROM AUTHOR]

This paper focus on developing a system which can monitors and warns parameters such as ultraviolet rays, gas, indicators of weather environment or the root conditions of explosion that outbreaks in or around a smart home within the connected range of any hazards which can be dangerous to people, thereby preparing and being proactive in possible situations. Through testing and practical use, the developed product has been operated stably, meeting the requirements of a monitoring and warning system for smart houses.