JournalTOCs

The Journal of Supercomputing
Journal Prestige (SJR): 0.407

Citation Impact (citeScore): 2
Number of Followers: 1

Hybrid journal (It can contain Open Access articles)
ISSN (Print) 1573-0484 - ISSN (Online) 0920-8542
Published by Springer-Verlag

[2468 journals]

A novel privacy-aware global infrastructure for ecological footprint
calculator based on the Internet of things and blockchain
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Over the past few decades, the harmful effects of industrial activity and consumer society have resulted in global warming. Governments and international organizations are looking for ways to monitor activities of individuals and companies to assess their ecological impact. Unfortunately, such an approach would easily be seen as a mass surveillance tool. This is why we are proposing in this paper a novel privacy-aware global infrastructure for ecological footprint calculator based on the Internet of things and blockchain. Indeed, we take advantage of the data collection capacity of the Internet of things, the anonymization provided by public key identification and encryption, and the immutability of blockchain to implement this global system. A three-stage approach was used to validate our architecture: modeling in Petri nets to verify that the infrastructure fulfills all the required missions, implementing the three central authorities with python to record parameters such as durations, and finally modeling in queuing networks to demonstrate stability. The blockchain and Internet of things parts are used in a purely abstract manner relying on standard concepts; therefore, we have not implemented them. The experiments have produced very promising results. We have shown that for the simplest form of queue modeling, the involved servers have a utilization rate that is close to \(50\%\) and that the overall waiting time remains below one minute.
  PubDate: 2023-12-09
Exploring autoregression patterns for automatic vessel type classification
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Automatic classification of vessel types in the maritime domain is one of the challenging problems due to the complexity of moving patterns in the ocean that are collected by the Automatic Identification System (AIS). In this study, we explore the usability of different patterns extracted from univariate and multivariate autoregressive modeling for classifying ship types. In order to assess the differentiation power of these features we apply different supervised machine learning classification algorithms and assess the performance of trajectory classification of four different vessel types. In addition, we study the effect of region specification for distinguishing the vessels. The proposed approach produced an accuracy of 86% which confirms that the features obtained from autoregression modeling can identify vessel types effectively. In addition, we demonstrate that the performance of classification can be enhanced further by considering the location of movement.
  PubDate: 2023-12-09
SNCL: a supernode OpenCL implementation for hybrid computing arrays
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Heterogeneous computing has been developing continuously in the field of high-performance computing because of its high performance and energy efficiency. More and more accelerators have emerged, such as GPU, FPGA, DSP, AI accelerator, and so on. Usually, the accelerator is connected to the host CPU as a peripheral device to form a tightly coupled heterogeneous computing node, and then, a parallel system is constructed by multiple nodes. This organization is computationally efficient, but not flexible. When new accelerators appear, it is difficult to join the system that has been built. At the hardware level, we create an array of accelerators and connect them to the existing system through a high-speed network. At the software level, we dynamically organize computing resources from various arrays to build a virtual heterogeneous computing node. This approach also includes a standard programming environment. Therefore, it is a more flexible, elastic, and scalable heterogeneous computing organization. In this paper, a supernode OpenCL implementation is proposed for hybrid parallel computing systems, in which virtual supernodes can be dynamically constructed between different computing arrays, and a standard OpenCL environment is implemented based on RDMA communication of high-speed interconnection, which can be combined with the system-level MPI programming environment, thereby realizing the large-scale parallel computing of the hybrid array. SNCL is compatible with existing MPI/OpenCL programs without the need for additional modifications. Experiments show that the runtime overhead of the supernode OpenCL environment is very low, and it is suitable for deploying applications with high computing density and large data scale between different arrays to utilize their computing power without affecting scalability.
  PubDate: 2023-12-08
A survey study on task scheduling schemes for workflow executions in cloud
computing environment: classification and challenges
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Several real-world scientific and industrial workflow applications adopt elastic and cost-efficient cloud services to fulfill their requirement. There are two stakeholders in the system, namely the user and the provider each of which tries to maximize its profit and at the same time minimize their possible overall costs. Since task scheduling algorithms for workflow executions determine which task should be possibly executed on what resources, they have a drastic impact on both the user’s quality of experiences and on the underlying resource utilization. Indeed, an efficient task scheduling algorithm can meet service-level agreement (SLA) for both sides. For the sake of the importance of the issue, this survey presents a subjective taxonomy on the task scheduling schemes in the literature for workflow executions in cloud computing environments to be a guideline for future improvement. It classifies the literature based on the proposed algorithms in the literature, objectives, stakeholders’ requirements, and evaluation metrics. This survey highlights research trends, challenges, research gaps, potential solutions, and future direction. It can also pave the way for further processing, improvement and strengthening of existing approaches or devising novel ones for interested researchers in the field of task scheduling problems.
  PubDate: 2023-12-07
An interactive multi-head self-attention capsule network model for aspect
sentiment classification
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The endpoint of aspect-level sentiment classification, a finely ground categorization task in sentiment analysis, is to gauge the polarity for various aspects in context. However, traditional attentional mechanisms still under-explore the relationship between aspect terms and context, and have difficulty in recognizing the overlapping features that arise when expressing multiple sentiment polarities to efficiently obtain deeper semantic representations. To solve these issues, we propose an interactive multi-head self-attention capsule network model (IMHSACap) for aspect sentiment classification. We design Local Context Mask to attenuate the influence of non-local contexts that are far away from the aspect terms, while expanding the influence of local contexts. Then the long-range intrinsic dependencies of global and local contexts are obtained by the interactive attention mechanism, which consists of two parts, Global2Local and Local2Global. The routing algorithm and activation function of the capsule network are optimized to improve the classification accuracy. Hence, experiments on three publicly available datasets are carried out to demonstrate that the IMHSACap model outperforms other baseline approaches for aspect sentiment classification.
  PubDate: 2023-12-07
A twofold bio-inspired system for mitigating SEUs in the controllers of
digital system deployed on FPGA
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Reconfigurable hardware, extensively employed in mission-critical digital applications like space and military electronics due to its adaptability, encounters the issue of soft errors, especially in control path elements, which could result in functional failure. Various system-level fault tolerance methodologies exist, and this paper implements a bio-inspired fault tolerance technique called evolvable hardware (EHW). The preferred implementation of the EHW system involves hosting the evolutionary algorithm on the processor alongside the reconfigurable hardware. However, this approach encounters delays in the intercommunication of the evolved circuit between the reconfigurable hardware and the processor. To address this issue, the paper proposes a two-tier architecture to achieve absolute fault mitigation in the controller. In this architecture, Tier-1 involves the digital implementation of the genetic algorithm on the reconfigurable hardware to mitigate errors in the controller, while Tier-2 focuses on mitigating errors occurring in Tier-1. The aim is to establish an absolute and self-resilient controller hardware to mitigate faults. The study simulates faults at the target circuit and genetic module as a proof of concept. The proposed two-tier single event upset (SEU) mitigation is deployed on Microsemi’s ProAsic3e FPGA (Field Programmable Gate Array), achieving an average efficiency of 91%. This efficiency is accompanied by ten times lesser resource utilization compared to traditional methodologies and a 30% accelerated speed when compared to hybrid evolvable systems.
  PubDate: 2023-12-07
A critical path task scheduling algorithm based on sequential failure
factor
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The effective utilization of task parallelism scheduling enables the exploitation of processing performance across multiple cores. Traditional serial scheduling algorithms are inadequate for achieving parallel task scheduling on multicore platforms. This study proposes a novel dynamic task scheduling algorithm called SEFAS. The algorithm aims to maximize task parallelism, leverage multicore performance, and reduce the duration of task set scheduling. The algorithm employs upward and downward ordering to calculate critical paths and categorize tasks into distinct levels. The introduction of node convergence factor and sequential failure factor facilitates the depiction of task parallelism and serves as a workaround for the limitations of the greedy algorithm. Experimental results demonstrate the enhanced adaptability of the proposed algorithm for computationally and communication-intensive tasks. Additionally, numerical comparisons reveal that the algorithm reduces the scheduling length by 35.0%, 93.7%, 17.1%, and 5% when compared to PLTSF, CPOP, CPMISF, and CPC, respectively.
  PubDate: 2023-12-07
Creation mechanism of new media art combining artificial intelligence and
internet of things technology in a metaverse environment
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The Metaverse is regarded as a brand-new virtual society constructed by deep media, and the new media art produced by new media technology will gradually replace the traditional art form and play an important role in the infinite Metaverse in the future. The maturity of the new media art creation mechanism must also depend on the help of artificial intelligence (AI) and Internet of Things (IoT) technology. The purpose of this study is to explore the image style transfer of digital painting art in new media art, that is, to reshape the image style by neural network technology in AI based on retaining the semantic information of the original image. Based on neural style transfer, an image style conversion method based on feature synthesis is proposed. Using the feature mapping of content image and style image and combining the advantages of traditional texture synthesis, a richer multi-style target feature mapping is synthesized. Then, the inverse transformation of target feature mapping is restored to an image to realize style transformation. In addition, the research results are analyzed. Under the background of integrating AI and IoT, the creation mechanism of new media art is optimized. Regarding digital art style transformation, the Tensorflow program framework is used for simulation verification and performance evaluation. The experimental results show that the image style transfer method based on feature synthesis proposed in this study can make the image texture more reasonably distributed, and can change the style texture by retaining more semantic structure content of the original image, thus generating richer artistic effects, and having better interactivity and local controllability. It can provide theoretical help and reference for developing new media art creation mechanisms.
  PubDate: 2023-12-06
Node utilization index-based data routing and aggregation protocol for
energy-efficient wireless sensor networks
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Data aggregation using the shortest path is a challenging issue in wireless sensor networks. Data aggregation slows the communication process and consumes many resources available to sensor nodes. So, while transmitting the aggregated data, it is crucial to identify the shortest and most reliable route for energy-efficient data aggregation from the source to the sink node. However, the nodes closer to the sink node are heavily utilized in multi-hop transmission and succumb to early death, which causes energy imbalance and an energy hole problem in the network. To overcome this problem, we propose an innovative node utilization index-based data routing and aggregation (NUIDRA) protocol. The NUIDRA protocol is designed and implemented in two phases. The first phase is to determine the shortest path from the source node to the sink node based on the amount of bandwidth utilized by each sensor node, minimum hop count, node’s residual energy, and data aggregation factor. In the second phase, the selected shortest path is used for data transmission and aggregation using the dynamic selection of the aggregator node. The choice of aggregator node is based on the node’s utilization index (UI) and adjacent node count from which it receives the data. The NUIDRA protocol is compared and analysed with I-LEACH and QADA protocols. The extensive simulation results show that in the proposed NUIDRA protocol, there is an increase in the average throughput of 70%, packet delivery ratio by 41.93%, and the average latency is reduced by 58.15% as compared to the I-LEACH protocol. Further, there is an increase in the average throughput of 24%, packet delivery ratio by 7.31%, and the average latency is reduced by 53.23% as compared to the QADA protocol for a data packet size of 512 bytes.
  PubDate: 2023-12-06
An effective and robust single-image dehazing method based on gamma
correction and adaptive Gaussian notch filtering
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The weather has a detrimental effect on outdoor vision systems and raises the probability of traffic crashes and road accidents. The scattering of atmospheric particles degrades outdoor images captured in poor weather conditions such as haze and fog. The reduced visibility has a significant impact on driving assistance systems designed for automatic vehicles. As a result, clear visibility is critical for outdoor computer vision systems. Image dehazing is one of the ill-posed problems because evaluating transmission depth is challenging. It is essential to estimate transmission depth with the greatest degree of accuracy. In order to estimate or optimize the transmission depth, this paper employs the adaptive Gaussian notch filter and the concept of gamma correction to recover the final scene radiance. The results of the experiments are assessed and compared both quantitatively and qualitatively with state-of-the-art techniques. The experimental results demonstrate that the proposed indicators ensure high consistency in qualitative and quantitative evaluation using six performance metrics: two blind assessment indicators (e, r), contrast gain \((C_{gain})\) , visual contrast measure (VCM), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM) and probability.
  PubDate: 2023-12-06
Analysis and efficient implementation of IEEE-754 decimal floating point
adders/subtractors in FPGAs for DPD and BID encoding
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract This paper proposes efficient implementations for addition/subtraction based on decimal floating point with Densely Packed Decimal (DPD) and Binary Integer Decimal (BID) encoding in FPGA devices. The designs use novel techniques based on the efficient utilization of dedicated resources in programmable devices. Implementations were made in Xilinx UltraScale+. For DPD adder/subtractor, they have computation times of 7.7 ns for Decimal32, 8.1 ns for Decimal64 and 8.5 ns for Decimal128. As for BID adder/subtractor, the computation time obtained is 13.5 ns for Decimal64. The proposed architecture achieves better computation times than related works. Compared to previous architectures, the proposed DPD implementation achieves 1.86 \(\times\) speedup and 47% better LUT occupation. Also, the BID adder/subtractor achieves 3 \(\times\) speedup and 5% less LUT occupation.
  PubDate: 2023-12-06
TSANET: transportation mode recognition model with global and local
spatiotemporal features
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Transportation mode recognition is an important and challenging problem in intelligent transportation systems. For decades, many data mining and deep learning methods have been proposed, but all these methods have some deficiencies and do not fully use temporal and spatial information hidden in raw GPS trajectory data. In this paper, we presented a new deep learning model called TSANET, which designs an attention mechanism to combine global and local spatiotemporal features. The input of the model is seven kinematic features, derived from the GPS trajectories and covers all information hidden in the GPS points. The model adopts TCN and ST-Block to extract global and local spatiotemporal features, respectively. In ST-Block, BiGRU is used to capture temporal features and DenseNet is applied to capture spatial features. In TCN, convolution operations are applied to obtain temporal and spatial information. Moreover, a dual-layer attention mechanism is developed to integrate and reconstruct these features by assigning different weights to global and local spatiotemporal features caught by TCN and ST-Block. By comprehensively considering global and local features, the model can greatly improve classification performance and recognition granularity, and with the further introduction of the attention mechanism, it can identify all transportation modes in a fine-grained manner. At last, the experiments have been conducted on the Geo1 and Geo2 datasets used by many researchers. The experimental results show that our model cannot only achieve the highest accuracy of 94.34 \(\%\) among all recent methods but also identify all seven transportation modes clearly, thus verifying the advantages and effectiveness of the model.
  PubDate: 2023-12-04
TentISSA-BPNN: a novel evaluation model for cloud service providers for
petroleum enterprises
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract To investigate how the petroleum industry evaluates and selects powerful cloud service providers, first, an evaluation index system including 25 indices such as scalability and private data protection is built. This index system can systematically examine the comprehensive strengths of cloud service providers. Aiming to solve the problems that the traditional expert evaluation method has high requirements on expert experience and is easily affected by subjective factors, a novel artificial intelligence evaluation model named TentISSA-BPNN is proposed. The objective evaluation ability of this model can be effectively used in evaluation research on cloud service providers for petroleum enterprises. In this model, the SSA algorithm is optimized by Tent chaotic mapping and adaptive inertia weight; an algorithm, TentISSA, that has good stability and fast convergence speed is designed and proposed; and the BPNN is improved with TentISSA to obtain more accurate evaluation results. To evaluate the performance of the TentISSA algorithm, nine unimodal and multimodal functions are selected in this paper to test the convergence accuracy. Then, seven models are selected as the control groups to validate the effectiveness and performance of the TentISSA-BPNN evaluation model proposed in this paper. Finally, the preprocessed data of the candidate cloud service providers are input into the trained neural network model proposed in this paper for evaluation. Based on the ranking of the evaluation scores, the comprehensive strengths of the cloud service providers are obtained to provide a decision-making reference for managers of petroleum enterprises in the process of choosing cloud service providers.
  PubDate: 2023-12-04
Green urban logistics path planning design based on physical network
system in the context of artificial intelligence
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract To address the issue of unreasonable route planning in modern logistics intelligent distribution, a comprehensive, efficient, and practical path planning method is provided by modern logistics intelligent distribution. This study takes a comprehensive multimodal transportation perspective between cities as its starting point. Firstly, a physical network system model describes logistics distribution’s fundamental physical network elements. Additionally, a logistics service network system model is established to depict the service network elements of logistics distribution. These two models are then combined. Building upon this foundation, variables such as risk and cost in logistics path planning are examined and considered, leading to the design of a logistics path planning model. Simulation analysis is employed in this study for data analysis. Analyzing section utilization and planning results reveals key information in logistics path planning. Segment 4 exhibits the lowest utilization rate at 46.2%, while Segment 7 demonstrates the highest utilization rate at 100%. This indicates that the model favors segment 7 as the primary path due to its superior physical facility elements and road conditions. By integrating section planning with cost and risk considerations, the optimal planning path is determined to be 1–3–6–7, encompassing sections 2, 7, and 10. This plan is based on the analysis results of the model and is considered the most economical and safe logistics path after comprehensive deliberation. The designed green logistics path planning model for the physical network system enables comprehensive planning of combined logistics routes, thereby enhancing the distribution path. The studied and designed green logistics path planning model of the physical network system allows for a comprehensive perspective in planning combined logistics traffic paths, resulting in improved distribution paths. This presents a novel and effective path-planning method for modern logistics intelligent distribution, accomplishing the research objectives and providing important references for subsequent related research.
  PubDate: 2023-12-03
On the construction of quality extended virtual backbones in wireless
sensor networks using cooperative communication
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Frequently, unit disk graphs (UDGs) are used as abstractions of wireless networks. In the paper, we study the problem of finding an extended minimum connected dominating set (E-MCDS) in a wireless network with cooperative communication (CC), which was proposed in Wu et al. (IEEE Trans Parallel Distrib Syst 17(8):851–864, 2006) and is NP-hard. We propose a two-phase centralized algorithm to construct an extended connected dominating set (ECDS) for a given UDG with CC. First, we construct a 2-hop extended dominating set (EDS) in the UDG. Then, we add some new nodes into the 2-hop EDS to make it connected. To obtain the performance ratio of this two-phase centralized algorithm, we first give an upper bound on the size of the 2-hop EDS. Using this upper bound, we prove that the size of the ECDS found by the centralized algorithm is no greater than \(27.192{\text {opt}}-3\) or \(160{\text {ECDS}}_{{\text {opt}}}-3\) , where opt and \({\text {ECDS}}_{{\text {opt}}}\) are the sizes of the minimum connected dominating set and the E-MCDS, respectively, in the UDG. Finally, a 160-approximate distributed algorithm based on the two-phase centralized algorithm is proposed, and its message and time complexities are analyzed, both of which are \(O(n^3)\) . Simulation results show that our work outperforms competing methods on average.
  PubDate: 2023-12-02
A mutual mean teacher framework for cross-domain aspect-based sentiment
analysis
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Aspect-based sentiment analysis is a fine-grained task that involves jointly extracting aspect terms and their corresponding sentiment polarities. However, due to the high cost of data labeling, obtaining annotated corpora for a specific target domain is often challenging. To address this issue, previous studies have attempted to transfer knowledge from a labeled source domain to the target domain by optimizing with pseudo-labels generated for the target domain data. These approaches allow both domains to be used to learn domain-invariant features. Nevertheless, such methods are susceptible to label noise, which hinders the extraction of domain-invariant features for the target domain. To mitigate the impact of error-prone pseudo-labels, we propose a mutual mean teacher framework for cross-domain aspect-based sentiment analysis. This framework generates pseudo-labels using a peer teacher network, thereby providing more reliable and robust pseudo-labels. Additionally, to develop a robust task classifier that performs well on both the source and target domains, we maximize the mutual information between the input token representations and the probability distributions of output labels, which helps prevent the model from making overly confident and incorrect predictions. Experiments conducted on ten different domain dataset pairs demonstrate that our proposed model exhibits competitive performance improvements compared to the current state-of-the-art model.
  PubDate: 2023-12-02
Application of deep learning and XGBoost in predicting pathological
staging of breast cancer MR images
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The methods of deep learning and traditional radiomics feature extraction were preliminarily discussed, and a multimodal data prediction model for breast cancer clinical stage was established. The MR images and clinical staging data of breast cancer were obtained from the official websites of the American Cancer Center TCGA and TCIA, respectively, with a total of 139 patient samples. The region of interest was delineated on the enhanced image of breast cancer MR, and then the feature extraction of radiomics and deep learning was performed, and 108 radiomics features and 1024 deep-learning features were extracted for each case. After feature screening and processing, clinical data were integrated, and a machine-learning model was used to predict clinical stage I and non-stage I. Results 26 radiomic features and 12 deep features related to staging were screened out by LASSO algorithm, and a classification model was constructed based on XGBoost machine learning. The patients were predicted with an accuracy rate of 80.00%, and the area under the curve of the receiver operating characteristic curve was 0.833. It is feasible to predict the clinical stage of breast cancer through radiomics and deep-learning feature extraction and machine-learning technology. The classification model based on multimodal data established by using machine-learning classifier can distinguish clinical stage I and non-stage I in breast cancer and have higher accuracy. This study confirms the feasibility and accuracy of combining data from different modalities to contribute to clinical staging prediction. The research contributions include demonstrating the superiority of deep-learning models for feature extraction and classification, as well as highlighting the potential of combining deep learning and traditional machine-learning algorithms for improved classification performance.
  PubDate: 2023-12-02
Anomaly detection in IOT edge computing using deep learning and
instance-level horizontal reduction
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract The increasing number of network attacks has led to the development of intrusion detection systems. However, these methods often face limitations such as high traffic flow data dimensions, which can reduce attack detection rates and noise sensitivity, affecting anomaly detection performance. This paper introduces a new model based on recurrent deep learning and instance-level horizontal reduction to detect anomalies and network attacks. The model uses nested sliding windows, which move with a specific step in the data and generate a different number of histogram outputs based on the type of anomaly in the data. Evaluation results on five databases show that the proposed model achieves a high accuracy of 99% in detecting different attacks, demonstrating the success of this new approach combined with deep recurrent neural networks in detecting anomalies.
  PubDate: 2023-12-02
Interoperability of heterogeneous Systems of Systems: from requirements to
a reference architecture
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Interoperability stands as a critical hurdle in developing and overseeing distributed and collaborative systems. Thus, it becomes imperative to gain a deep comprehension of the primary obstacles hindering interoperability and the essential criteria that systems must satisfy to achieve it. In light of this objective, in the initial phase of this research, we conducted a survey questionnaire involving stakeholders and practitioners engaged in distributed and collaborative systems. This effort resulted in the identification of eight essential interoperability requirements, along with their corresponding challenges. Then, the second part of our study encompassed a critical review of the literature to assess the effectiveness of prevailing conceptual approaches and associated technologies in addressing the identified requirements. This analysis led to the identification of a set of components that promise to deliver the desired interoperability by addressing the requirements identified earlier. These elements subsequently form the foundation for the third part of our study, a reference architecture for interoperability-fostering frameworks that is proposed in this paper. The results of our research can significantly impact the software engineering of interoperable systems by introducing their fundamental requirements and the best practices to address them, but also by identifying the key elements of a framework facilitating interoperability in Systems of Systems.
  PubDate: 2023-12-02
AAR:Attention Remodulation for Weakly Supervised Semantic Segmentation
- Free pre-print version: Loading...
  Rate this result: What is this?
  
  Please help us test our new pre-print finding feature by giving the pre-print link a rating.
  A 5 star rating indicates the linked pre-print has the exact same content as the published article.
  
  Abstract: Abstract Weakly Supervised Semantic Segmentation is a crucial task in computer vision. However, existing methods that utilize Class Activation Maps (CAMs) with classification tasks can only identify a small part of the region. To address this limitation, we propose a novel Attention Activation Remodulation (AAR) scheme that leverages traditional CAMs and the remodulation branch to obtain weighted CAMs for recalibrated supervision. The AAR scheme re-arranges important features’ distribution from the channel and space perspectives, which regulates segmentation-oriented activation responses. In addition, we propose a Feature Pixel Extraction Module (FPEM) that utilizes contextual information to improve pixel prediction. Furthermore, the proposed scheme can be combined with other methods to improve overall performance. Extensive experiments on the PASCAL VOC 2012 dataset demonstrate the effectiveness of the AAR mechanism and FPEM module.
  PubDate: 2023-12-02