Topological spin structures with transformable shapes may have potential implications on data storage and computation. Here, we demonstrate that a square cellular skyrmion on an artificial grid pinning pattern can be ...
详细信息
Many services that are taken for granted in smart cities are not even remotely available in dislocated areas, i.e. "smart territories". With the aim to offer a practical and secure way to transport data in s...
Many services that are taken for granted in smart cities are not even remotely available in dislocated areas, i.e. "smart territories". With the aim to offer a practical and secure way to transport data in such constrained scenarios, we focus on the problem of incentivizing to Data Mules, i.e. devices dedicated to enable communication even in the absence of the Internet. We combine decentralized technologies and State-Channels to verify the correct behavior of participants in an offline scenario.
Hyperledger Fabric stands as a leading framework for permissioned block-chain systems, ensuring data security and audit-ability for enterprise applications. As applications on this platform grow, understanding its com...
详细信息
ISBN:
(数字)9798350327939
ISBN:
(纸本)9798350327946
Hyperledger Fabric stands as a leading framework for permissioned block-chain systems, ensuring data security and audit-ability for enterprise applications. As applications on this platform grow, understanding its complex configuration concerning various block-chain parameters becomes vital. These configurations significantly affect the system’s performance and cost. In this research, we introduce a Stochastic Petri Net (SPN) model to analyze Hyper-ledger Fabric’s performance, considering variations in block-chain parameters, computational resources, and transaction rates. We provide case studies to validate the utility of our model, aiding block-chain administrators in determining optimal configurations for their applications. A key observation from our model highlights the block size’s role in system response time. We noted an increased mean response time, between 1 to 25 seconds, due to variations in transaction arrival rates.
In the latest days, study into the development of intelligent technologies has proven valuable, contributing to attempts to improve the quality of human existence. Smart glass is one of the intelligent wearable device...
In the latest days, study into the development of intelligent technologies has proven valuable, contributing to attempts to improve the quality of human existence. Smart glass is one of the intelligent wearable devices that can be used for various purposes, including healthcare monitoring, fall detection, sleep tracking, and human activity recognition (HAR). Smart-phones and smartwatches are the primary wearables utilized in sensor-based HAR to collect human motions for training recognition models based on physical movement. These wearable tools, nevertheless, are more intrusive than smart glasses. Using IMU sensor data acquired via smart glasses, we investigate deep learning algorithms for detecting people's activities of daily living (ADL). This work proposes a hybrid deep neural network that automatically extracts spatial-temporal information from raw data to enhance identification $\mathbf{p}$ erformance. We performed tests to evaluate deep learning models using a publically available benchmark dataset, UCA-EHAR, which included IMU sensor data from multiple ADL from smart eyewear. The recommended CNN-LSTM model achieved the best effectiveness with the highest F1-score of 93.20%, as determined by experimental findings.
The prevalence of intelligent wearable devices has significantly increased, offering advantages to individuals of all age groups. In the field of human activity recognition (HAR) research, wearable sensor data plays a...
The prevalence of intelligent wearable devices has significantly increased, offering advantages to individuals of all age groups. In the field of human activity recognition (HAR) research, wearable sensor data plays a crucial role in classifying various human actions. Sensor-based assessments are commonly employed in hazard analysis and risk assessments. However, traditional machine learning (ML) algorithms have faced challenges when applied to sports activities due to their unpredictable nature. These challenges have prompted the need for manual feature extraction, limiting the efficacy of ML methodologies in data categorization. To address this, our study introduces ResNet, a deep residual network, for effectively categorizing sports activities using wearable sensor data. To evaluate our proposed approach, we utilized a publicly available benchmark dataset consisting of sensor data collected from the dominant wrist, neck, and thigh of each study participant. The experimental results demonstrated the exceptional performance of the ResNet model, achieving an impressive accuracy rate of 99.83% in recognizing sport activities. Additionally, we compared the efficacy of our suggested model with five fundamental deep learning models, namely CNN, LSTM, BiLSTM, GRU, and BiGRU. The comparative analysis revealed that the ResNet model outperformed the other deep learning models, demonstrating superior performance in sport activity recognition.
作者：
Skala, VaclavUniversity of West Bohemia
Faculty of Applied Sciences Department of Computer Science and Engineering Univerzitni 8 PlzenCZ 306 14 Czech Republic
This paper presents a new approach to computation of geometric continuity for parametric bi-cubic patches, based on a simple mathematical reformulation which leads to simple additional conditions to be applied in the ...
详细信息
Many algorithms used are based on geometrical computation. There are several criteria in selecting appropriate algorithm from already known. Recently, the fastest algorithms have been preferred. Nowadays, algorithms w...
详细信息
Human activity recognition (HAR) has become a hot topic in artificial intelligence research due to the rapid development of smart wearable technologies. The goal of HAR is to accurately identify human actions using va...
Human activity recognition (HAR) has become a hot topic in artificial intelligence research due to the rapid development of smart wearable technologies. The goal of HAR is to accurately identify human actions using various data sources, such as video, images, and sensor data from wearable devices. Recent research in HAR has achieved promising results using learning-based methods, especially deep learning techniques. However, achieving state-of-the-art results remains a challenge for researchers. This study proposes a new approach to HAR that uses deep learning to classify human activities from smartwatch sensor data. We propose the use of a one-dimensional deep pyramidal residual network (1D-PyramidNet) for accurate human action identification. We evaluate the performance of our model against baseline models using the DHA dataset, a benchmark dataset for HAR that includes wristwatch sensor data for 11 complex human activities. The experimental results show that our 1D-PyramidNet model outperforms the baseline models, including CNN, LSTM, BiLSTM, GRU, and BiGRU. This confirms that the use of 1D-PyramidNet can improve the identification capabilities of HAR systems, achieving a maximum accuracy of 96.64%.
The use of ensemble techniques is widely recognized as the most advanced approach to solving a variety of problems in machine learning. These strategies train many models and combine the results from all of those mode...
详细信息
The use of ensemble techniques is widely recognized as the most advanced approach to solving a variety of problems in machine learning. These strategies train many models and combine the results from all of those models in order to enhance the predictive performance of a single model. During the period of the last several years, the disciplines of artificial intelligence, pattern recognition, machine learning, neural networks, and data mining have all given a considerable consideration to the concept of ensemble learning. Ensemble Learning has shown both effectiveness and usefulness across a broad range of problem domains and in significant real-world applications. Ensemble learning is a technique thatinvolves the construction of many classifiers or a group of base learneis and the merging of their respective outputs in order to decrease the total variance. When compared to using only one classifier or one base learner at a time, the accuracy of the results achieved by combining numerous dassifiers or the set of base learners is greatly improved. It has been shown that the use of ensemble methods may increase the predicted accuracy of machine learning models for a range of tasks, including classification, regression, and the identification of outliers. This study will discuss about ensemble machine learning techniques and its various methods such as bagging, boosting, and stacking. finally, all the factors involved in bagging, boosting, and stacking are compared.
The Internet of Vehicles (IoV) is an architecture of the intelligent transportation system that combines automotive, transportation, and information exchange to increase road safety. The categorization of roadways not...
The Internet of Vehicles (IoV) is an architecture of the intelligent transportation system that combines automotive, transportation, and information exchange to increase road safety. The categorization of roadways not only improves the passenger's comfort and safety but also provides autonomous cars with safe navigation paths. In this paper, we present an identification method based on movement data from smart glasses (electroocu-lography, acceleration, and angular velocity) to categorize four kinds of roads often experienced: highway, city road, highway, undeveloped region, and housing estate. We developed a deep pyramidal residual network that automatically recovers spatial-temporal data and efficiently identifies road kinds. We performed experiments to evaluate deep learning models utilizing a publicly available benchmark dataset, including sensor data acquired from intelligent eyewear. Experimentally, we discovered that the suggested 1D-PyramidNet model obtained the most incredible interpretation with the most increased accuracy (92.23%) and outperformed all other deep learning models.
暂无评论