With the accelerated development of the equipment under test towards informatization and systemization, higher requirements are put forward for the construction of the test capacity of the equipment under test. It is ...
详细信息
Information systems depend on security mechanisms to detect and respond to cyber-attacks. One of the most frequent attacks is the Distributed Denial of Service (DDoS): it impairs the performance of systems and, in the...
详细信息
ISBN:
(数字)9783031530258
ISBN:
(纸本)9783031530241;9783031530258
Information systems depend on security mechanisms to detect and respond to cyber-attacks. One of the most frequent attacks is the Distributed Denial of Service (DDoS): it impairs the performance of systems and, in the worst case, leads to prolonged periods of downtime that prevent business processes from running normally. To detect this attack, several supervised Machine Learning (ML) algorithms have been developed and companies use them to protect their servers. A key stage in these algorithms is feature pre-processing, in which, input data features are assessed and selected to obtain the best results in the subsequent stages that are required to implement supervised ML algorithms. In this article, an innovative approach for feature selection is proposed: the use of Visibility Graphs (VGs) to select features for supervised machine learning algorithms used to detect distributed DoS attacks. The results show that VG can be quickly implemented and can compete with other methods to select ML features, as they require low computational resources and they offer satisfactory results, at least in our example based on the early detection of distributed DoS. The size of the processed data appears as the main implementation constraint for this novel feature selection method.
With the explosive increase in mobile data volume, traditional cloud platforms can no longer meet the real-time requirements of edge devices. In this context, some scholars have proposed deploying a middle cloud, also...
详细信息
ISBN:
(纸本)9798350376784;9798350376777
With the explosive increase in mobile data volume, traditional cloud platforms can no longer meet the real-time requirements of edge devices. In this context, some scholars have proposed deploying a middle cloud, also known as an edge cloud. The data storage and processing capabilities of the edge cloud are far inferior to those of the center cloud, thus it cannot efficiently handle large-scale edge data samples. Moreover, traditional wireless communication pursues high data rates or reliable transmission, with data bits being transmitted indiscriminately to the network edge, resulting in high redundancy. Therefore, reducing the scale of edge data samples and eliminating high redundant data samples has become an urgent issue to be addressed. This paper mainly focuses on the edge service side and proposes a method for reducing data sample redundancy based on data-importance analysis, which includes three steps: firstly, clustering the original edge data samples to obtain important classes;Secondly, calculating the similarity of samples within each important class;And thirdly, screening the samples. We trained SVM, KNN, and DT classification models for the Iris dataset using the proposed method. The average accuracy of the DT and SVM models has increased by 1.192% and 3.432% respectively, while it has almost no impact on the accuracy of KNN. The experiment shows that removing highly redundant samples can still preserve the features of model training, and even improve the accuracy of the model.
Validating intelligent algorithms within the realm of Autonomous driving Domain Controllers (ADC) using virtual simulation platforms has become a foundational industry paradigm. Modeling the perception system of auton...
详细信息
The demand for multimedia content imposes considerable strain on IP technology, particularly in terms of latency. Named data Networking (NDN) emerges as a future networking solution, characterized by a content-centric...
详细信息
ISBN:
(纸本)9798350351774;9798350351767
The demand for multimedia content imposes considerable strain on IP technology, particularly in terms of latency. Named data Networking (NDN) emerges as a future networking solution, characterized by a content-centric approach featuring storage capabilities on routers. However, challenges persist in effectively managing the limited storage capacity of routers, necessitating the development of efficient content replacement strategies. Traditional caching replacement algorithms often struggle to adapt to the dynamic nature and popularity of content across the network. Previous studies have proposed leveraging machine learning (ML) to enhance caching replacement strategies, ensuring eviction processes target only unpopular content. These endeavors have demonstrated promising results in enhancing router storage efficiency. Nonetheless, a critical inquiry arises regarding integrating ML models within the network: Should the ML model be deployed directly on routers or exclusively at the producer level? Bridging the gap between theoretical research and practical implementation is imperative. This study is dedicated to implementing ML algorithms on routers within the NDN. It shows how a simple K-Nearest Neighbors (KNN) machine learning algorithm combined with the Least Recently Used (LRU) caching strategy could greatly improve the process of replacing NDN cache when embedded in all NDN routers. The proposed KNN-LRU model works better than common algorithms like LRU, FIFO, and LFU in terms of cache hit ratio, latency, and link load when validated using the ICARUS caching simulator.
Fuzzy dataprocessing enables data enrichment and increases data interpretation in industrial environments. In the cloud-based IoT data ingestion pipelines, fuzzy dataprocessing can be implemented in several location...
详细信息
Virtual Reality (VR) environments are used as a way to let humans experiment with algorithms, techniques, and situations in various application areas, including emergency management, serious games, smart manufacturing...
详细信息
ISBN:
(纸本)9798350375039;9798350375022
Virtual Reality (VR) environments are used as a way to let humans experiment with algorithms, techniques, and situations in various application areas, including emergency management, serious games, smart manufacturing, and precision agriculture. They are especially relevant when experiments in the real world may be harmful to human operators. As VR environments are the closest possible faithful replicas of real environments, many recent works focus on the employment of such tools as a means to generate synthetic datasets that can be used for training machine and deep learning models, especially in situations where obtaining real datasets can be difficult. In this paper, we introduce a strategy to generate a dataset for pose estimation in the challenging scenario of precision agriculture. Finally, the quality of the generated dataset was evaluated.
Extractive text summarization is a well-studied downstream task of natural language processing that aims to select sentences as a summary of the document's critical information. For the systems built based on pre-...
详细信息
ISBN:
(纸本)9781665488679
Extractive text summarization is a well-studied downstream task of natural language processing that aims to select sentences as a summary of the document's critical information. For the systems built based on pre-trained language models, the large volume of parameters can cause the model performance gets significantly degradation when the training data is insufficient. However, acquiring high-quality labeled data is a time-consuming and laborious task. Previous works mainly focus on accuracy, and few of them pay attention to the robustness of the model. Hence training a robust and high-quality model is the concern of this work. In this work, we propose RBPSum, a robust extractive summarization model based on the pre-trained language model. Through the experiments, we find that under the situation of restricted data size, sentence position information plays a critical role in extractive summarization. In terms of ROUGE metrics, our model outperforms the previous state-of-the-art approaches when using the entire training set, and around a third of the training set produces competitive results.
Every single day, users are increasing as the world has explored technology where data is producing more and more big data and cloud technologies are the reason behind the scene because big data and cloud computing ar...
详细信息
With the development of the Internet and the large-scale digital music industry, the acquisition and listening of music are presented to users in a more convenient way. How to find the music that users like to hear fr...
详细信息
暂无评论