Given the critical role of graphs in real-world applications and their high-security requirements, improving the ability of graph neural networks (GNNs) to detect out-of-distribution (OOD) data is an urgent research p...
Frequently, individuals undergo specific episodes of mental health challenges throughout their lifetime. But the COVID pandemic has triggered a surge in mental health disorders arising from isolation, monotonous routi...
Frequently, individuals undergo specific episodes of mental health challenges throughout their lifetime. But the COVID pandemic has triggered a surge in mental health disorders arising from isolation, monotonous routines, demanding workloads, financial disparities, and disruptions to daily schedules. Furthermore, the global pandemic has induced constant anxiety and stress. Beyond the pandemic, the competition and intense pressure of the modern world impact mental health. Access to advanced mental health solutions and the necessary familiarity remain limited for most of the population. Given the integration of technology into daily life, diverse remedies, including mobile and web applications, have emerged to tackle the escalating challenge of mental health disorders. This study proposes an accessible and cost-effective approach that employs machine learning to detect stress levels and discern user emotions from journal entries and facial expressions while integrating self-journaling, video recommendations, and visual content generation to stimulate positive emotions and relieve stress.
Satellite detection refers to the identification and recognition of satellites in satellite images. This task is crucial for various applications, including space surveillance, satellite tracking, and space debris mon...
Satellite detection refers to the identification and recognition of satellites in satellite images. This task is crucial for various applications, including space surveillance, satellite tracking, and space debris monitoring. Several studies have been conducted to develop methods and techniques for satellite detection. The intricacy and scale of satellite images make satellite detection incredibly challenging. The process of manual identification and classification necessitates a significant amount of time. Recently, Deep Learning (DL) has emerged as a highly promising approach in computer Vision (CV) tasks. This study evaluates the effectiveness of three widely employed DL-based object detection algorithms, specifically YOLOv4, YOLOv5, and YOLOv8, within the field of satellite detection. The algorithms have undergone training utilizing a dataset of satellite photos comprising both satellites and debris. The evaluation of the algorithm’s performance involved the utilization of broadly accepted metrics, including Precision, Recall, Mean Average Precision (mAP), Box Loss, and Class Loss. The findings indicated that all three algorithms exhibited a high level of accuracy in detecting and precisely determining the location of objects inside satellite data. The performance of YOLOv5 and YOLOv8 surpassed that of YOLOv4 across various measures, with a notable advantage observed in detecting tiny objects. Overall, this study demonstrates the capabilities of DL algorithms in the context of satellite detection and classification techniques.
Semi-supervised learning (SSL) is a promising solution for the problem of insufficient medical labeled data. However, it is still a challenging task to segment transvaginal ultrasound (TVUS) images because of their po...
详细信息
ISBN:
(数字)9798350350548
ISBN:
(纸本)9798350350555
Semi-supervised learning (SSL) is a promising solution for the problem of insufficient medical labeled data. However, it is still a challenging task to segment transvaginal ultrasound (TVUS) images because of their poor contrast and speckle noise. In this study, we propose Multi-StudentNet, a semi-supervised deep-learning framework for endometrial segmentation in TVUS images. This method employs multiple student models to effectively integrate labeled and unlabeled data, improving the reliability and performance of segmentation while significantly reducing the need for extensive manual labeling. By leveraging the weights of multiple models, our framework facilitates feature sharing, which enhances model robustness. Specifically, Multi-StudentNet achieves a Dice coefficient (DSC) of 0.81 and a specificity of 0.99. Our multi-student models consistently outperform single-student models across various conditions, including normal, polyp, and cancer cases. Extensive experiments demonstrate that Multi-StudentNet achieves state-of-the-art performance in both accuracy and robustness.
Rockets and missiles fired from a tube usually have aerodynamic surfaces that are packed when the rocket is in the tube. One of the common fins folding solutions is the wrap-around fins. The wrap-around fins are usual...
详细信息
Traditional anti-jamming techniques like spread spectrum, adaptive power/rate control, and cognitive radio, have demonstrated effectiveness in mitigating jamming attacks. However, their robustness against the growing ...
详细信息
Modern Internet of Thing (IoT) applications generate enormous amount of data. To prototype the incoming data from such applications, data-driven machine learning has emerged as a viable method which can help to develo...
详细信息
By the beginning of 2020, the world woke up to a global pandemic that changed people’s everyday lives and restrained their physical contact. During those times Social Media Platforms (SMPs) were almost the only mean ...
详细信息
The Internet of Things (IoT) involves complex, interconnected systems and devices that depend on contextsharing platforms for interoperability and information exchange. These platforms are, therefore, critical compone...
详细信息
ISBN:
(数字)9798331510183
ISBN:
(纸本)9798331510190
The Internet of Things (IoT) involves complex, interconnected systems and devices that depend on contextsharing platforms for interoperability and information exchange. These platforms are, therefore, critical components of real-world IoT deployments, making their security essential to ensure the resilience and reliability of these “systems of systems.” In this paper, we take the first steps toward systematically and comprehensively addressing the security of IoT context-sharing platforms. We propose a framework for threat modelling and security analysis of a generic IoT context-sharing solution, employing the MITRE ATT&CK framework. Through an evaluation of various industry-funded projects and academic research, we identify significant security challenges in the design of IoT context-sharing platforms. Our threat modelling provides an in-depth analysis of the techniques and sub-techniques adversaries may use to exploit these systems, offering valuable insights for future research aimed at developing resilient solutions. Additionally, we have developed an open-source threat analysis tool that incorporates our detailed threat modelling, which can be used to evaluate and enhance the security of existing context-sharing platforms.
Social commerce uses a variety of social networking sites like Facebook, Instagram, Twitter, Pinterest, and business Whatsapp as a platform to promote products and services. Since advertising, a product, or service in...
详细信息
ISBN:
(数字)9798350384369
ISBN:
(纸本)9798350384376
Social commerce uses a variety of social networking sites like Facebook, Instagram, Twitter, Pinterest, and business Whatsapp as a platform to promote products and services. Since advertising, a product, or service in s-comm will reach as high as a possible number of users on networking sites. Ads will spread virally by sharing among friends and friends of friends. So that wider reachability is possible and shortly will get the customer for our business. The success of an s-comm campaign is measured by the degree to which consumers interact with the sellers through review comments, likes, and shares between friends. Customer behavior analysis is needed for s-comm marketing because social commerce can significantly boost economic growth. The sentiment analysis of customer recommendations was investigated using a dataset of women's s-comm apparel hashtags and reviews. So for such reviews are experimented with different machine learning algorithms (Decision tree, SVM, Random Forest, KNN, Naïve Bayes and LSTM). These models handled the common English reviews and compared the metrics (Accuracy, Sensitivity and Specificity) with one another. Instead of working with individual model, this paper introduces a novel hybrid approach that integrates CNN and Ada Boost LSTM algorithms to provide better results in sentiment analysis when compared to other machine learning algorithms mentioned above. While considering the hash tag with common review comments the hybrid system is out performed. The comparison analysis is done for performance matrices namely accuracy of 98%, specificity of 98.1%, and sensitivity of 98.1%. In the s-commerce transaction industry, this hybrid approach will help to create a deeper understanding of consumer sentiment analysis and psychologically attract clients.
暂无评论