Showing papers in "Future Generation Computer Systems in 2021"

TL;DR: This paper aims to provide a comprehensive study concerning FL’s security and privacy aspects that can help bridge the gap between the current state of federated AI and a future in which mass adoption is possible.

TL;DR: An Attention-based Bidirectional CNN-RNN Deep Model (ABCDM) is proposed that achieves state-of-the-art results on both long review and short tweet polarity classification and is evaluated on sentiment polarity detection.

TL;DR: A novel healthcare monitoring framework based on the cloud environment and a big data analytics engine is proposed to precisely store and analyze healthcare data, and to improve the classification accuracy.

TL;DR: This article has proposed an ensemble classification model for detection of the fake news that has achieved a better accuracy compared to the state-of-the-art.

TL;DR: In this article, the authors investigated the drone-based systems, COVID-19 pandemic situations, and proposed an architecture for handling pandemic situation in different scenarios using real-time and simulation-based scenarios.

TL;DR: A three-module framework, named “Ontology-Based Privacy-Preserving” (OBPP) is proposed to address the heterogeneity issue while keeping the privacy information of IoT devices, and can be widely applied to smart cities.

TL;DR: The indoor scene semantic segmentation model constructed in this paper not only has good performance and high efficiency, but also can segment the contours of different scale objects clearly and adapt to the indoor uneven lighting environment.

TL;DR: Numerical results illustrate that the proposed schemes can effectively prevent data poisoning attacks and improve the privacy protection of model updates for secure and privacy-preserving traffic flow prediction.

TL;DR: An identity-based proxy aggregate signature (IBPAS) scheme is proposed to improve the efficiency of signature verification, as well as compress the storage space and reduce the communication bandwidth.

TL;DR: This research proposes a method to conduct calculations in a collaborative way to alleviate the huge computing pressure caused by the single mobile edge server computing mode as the amount of data increases.

TL;DR: It can be concluded that the prediction of the stock price trend prediction model of BP algorithm neural network is better than that of the deep learning fuzzy algorithm prediction model.

TL;DR: A container scheduling system that enables serverless platforms to make efficient use of edge infrastructures and a method to automatically fine-tune the weights of scheduling constraints to optimize high-level operational objectives such as minimizing task execution time, uplink usage, or cloud execution cost is presented.

TL;DR: A novel approach DeepAMD to defend against real-world Android malware using deep Artificial Neural Network (ANN) has been adopted including an efficiency comparison of DeepAMD with conventional machine learning classifiers and state-of-the-art studies based on performance measures such as accuracy, recall, f-score, and precision.

TL;DR: A novel high-efficient approach is proposed named DIDDOS to protect against real-world new type DDoS attacks using Gated Recurrent Unit (GRU) a type of Recurrent Neural Network (RNN).

TL;DR: This paper proposes a multi-dimensional feature fusion and stacking ensemble mechanism (MFFSEM), which can detect abnormal behaviors effectively and significantly outperforms the basic and meta classifiers adopted in the method.

TL;DR: A cluster analysis from the keyword perspective was carried out to obtain emerging trends and frontiers of blockchain research and showed that future research should concentrate on management, blockchain technology, energy, machine learning, and smart home.

TL;DR: Experimental results indicate that the proposed algorithm can achieve the stability and efficiency of task scheduling and effectively improve the throughput of the cloud computing system.

TL;DR: This paper addresses the challenge of automatic identification of aggression detection on tweets of cyber-troll dataset by deploying Multilayer Perceptron and experimenting on state-of-the-art combination of CNN-LSTM and CNN-BiL STM in deep neural network, both perform well.

TL;DR: Zhang et al. as mentioned in this paper proposed a BiLSTM-based attention mechanism with a dilated convolutional neural network (DCNN) that selectively focuses on effective features in the input frame to recognize the different human actions in the videos.

TL;DR: A novel hybrid approach based on Dynamic Malware Analysis, Cyber Threat Intelligence, Machine Learning (ML), and Data Forensics is proposed which is able to reduce the security issues which were neglected by existing outdated reputation engines.

TL;DR: This research on the combination of a CPS and artificial intelligence in the construction industry can provide a theoretical basis and practical reference materials for the development of the intelligent construction industry.

TL;DR: A Bayesian game theory-based solution to empower service provider to maximize the social welfare by employing incentives and pricing rules on the users of a network and proposes Bayesian pricing and auction mechanism to achieve Bayesian Nash Equilibrium points in different scenarios.

TL;DR: A Deep Graph neural network-based Spammer detection (DeG-Spam) model under the perspective of heterogeneous cyberspace is proposed and representations for occasional relations and inherent relations are separately modelled.

TL;DR: An alternative task scheduling technique for IoT requests in a cloud-fog environment based on a modified artificial ecosystem-based optimization (AEO), called AEOSSA, which demonstrates the high ability to tackle the task scheduling problem and perform better than other methods according to the performance metrics.

TL;DR: A deep learning model that integrates S-Mask R-CNN and Inception-v3 in the ultrasound image-aided diagnosis of prostate cancer in this paper has higher accuracy than that of the doctor’s manual diagnosis and other detection methods.

TL;DR: A new methodological framework is proposed that can make future investigations in this research field easier, coherent, and uniform in the context of anomaly detection in an MIoT, and the so-called "forward problem” and "inverse problem" are defined.

TL;DR: The proposed multiple improved residual network (MIRN) super-resolution reconstruction method can better reconstruct the details and textures of images and avoid the over-smoothing of medical images after iteration, and the performance of the proposed algorithm is revealed to be better than that of existing state-of-the-art methods.

TL;DR: This paper reviews and systematizes the state-of-the-art solutions that address both DoS and DDoS attacks in SDNs through the lenses of intrinsic and extrinsic approaches, and surveys the different approaches and tools adopted to implement the revised solutions.

TL;DR: This paper augments the existing security framework with an Autonomous Response Controller (ARC) that uses the authors' quantitative Hierarchical Risk Correlation Tree (HRCT) that models the paths an attacker can traverse to reach certain goals and measures the financial risk that the CPS assets face from cyber-attacks.

TL;DR: The experimental results show that the proposed DEQP2 model provides measurable privacy preservation without significantly reducing the accuracy.