Search

In the era of 5G/B5G, computing-intensive, delay-sensitive applications such as virtual reality inevitably bring huge amounts of data to the network. In order to meet the real-time requirements of applications, Mobile Edge Computing (MEC) pushes computing resources and data from the centralized cloud to the edge network, providing users with computing offload technology. However, the mismatch between the great computing requirements of computing-intensive tasks and the limited computing power of a single edge server poses a great challenge to computing offload technology. In this paper, a multi-agent cooperation mechanism for MEC and a routing mechanism based on deep reinforcement learning (DRL) are proposed. First of all, a multi-agent cooperation mechanism is proposed to realize the cooperative processing of computing-intensive and delay-sensitive applications, and the task unloading decision-making problem based on multi-agent cooperation is studied. Secondly, the cooperative processing of tasks by multi-agents involves data transmission. Considering the real-time requirements of tasks, this paper proposes an intelligent routing mechanism based on DRL to plan the optimal routing path. Finally, the simulation implementation and performance evaluation of the multi-agent cooperation mechanism and routing mechanism for MEC are carried out. The experimental results show that the intelligent routing mechanism based on DRL and graph neural network is superior to the comparison mechanism in terms of network average delay, throughput and maximum link bandwidth utilization. At the same time, the superiority of graph neural network in model generalization is verified on a new network topology National Science Foundation (NSF) Net. The results of route optimization are applied to the multi-agent cooperation mechanism, and the experimental results show that the mechanism is superior to the comparison scheme in terms of task success rate and average task response delay. The combination of these two mechanisms well solves the problem that it is difficult to deal with computing-intensive and delay-sensitive applications in mobile edge computing because of its limited resources. [ABSTRACT FROM AUTHOR]

The identification, prediction, management, and control of economic and financial systems render extremely challenging tasks, which require comprehensively integrating the knowledge of different expert systems. Ontology, as a state-of-the-art knowledge modeling technique, has been extensively applied in the domain of economics and finance. However, due to ontology engineers' subjectivity, ontology suffers from the heterogeneity issue, which hampers the co-operation among the intelligent expert system based on them. To address this issue, ontology matching for finding heterogeneous concept pairs between two ontologies has been rapidly developed. It is difficult to find the perfect ontology alignment that satisfies the needs of all decision-makers. Therefore, Multi-Objective Evolutionary Algorithm, such as Non-dominated Sorting Genetic Algorithm (NSGA-II), attracts many researchers' attention. However, when facing large-scale ontology matching problems, NSGA-II tends to fall into local optimal solutions due to the large search space. To effectively address this drawback, we model the large-scale ontology problem as a nonlinear optimization problem, and propose an Adaptive Co-Evolutionary NSGA-II (ACE-NSGA-II) to deal with it. Compared with NSGA-II, ACE-NSGA-II introduces a co-evolutionary mechanism to increase the diversity of populations in order to decrease the probability of premature convergence. In particular, ACE-NSGA-II uses an adaptive population maintenance strategy to assign more resources toward the dominant ones in order to improve the solution efficiency for solving large-scale ontology matching. The experiment utilizes the Ontology Alignment Evaluation Initiative (OAEI)'s benchmark and anatomy track to test the effectiveness of ACE-NSGA-II, and the resulting experiment demonstrated that compared to NSGA-II and OAEI's participants, ACE-NSGA-II is able to find better alignment. [ABSTRACT FROM AUTHOR]

Summary: Information‐Centric Networking (ICN) has been accepted to overcome some weaknesses of the current Internet architecture, showing that "what is being exchanged" is more important than "who are exchanging information." Given the inadequate considerations on Quality of Service (QoS) and energy saving in ICN routing, we propose in this paper a routing algorithm to enhance the two aspects. At first, on one hand, Cauchy distribution is used as a fuzzy model to evaluate users' QoS requirements, such as bandwidth, delay, and error rate; on the other hand, we formulate energy saving problem to evaluate the green quality of routing algorithm. Then, we design a link selection approach by considering QoS and energy saving, which belongs to a multi‐objective decision problem resolved by intelligent drops algorithm. Finally, we implement the proposed algorithm and compare it with the famous adaptive forwarding mechanism in terms of some significant metrics, and the experimental results reveal that the proposed algorithm is more efficient. [ABSTRACT FROM AUTHOR]

With the promotion and development of oversized data technology, many data analysis platforms based on super large data storage and computing frameworks have emerged in the industry. While the platforms with oversized economic data analysis combined with machine learning models are still relatively lacking. And oversized data also brings a new problem, that is the security of economic development. It is an important and difficult task to analyse and detect risks from oversized economic data. Based on machine learning, data analysis, economic market and other multidisciplinary fields, this paper proposes a machine learning method, which is a genetic algorithm (GA) based feature selection method: FSGA. This method abstracts every possible feature selection result into an individual in GA, generates a population through genetic operation, and measures the merits of the individual through fitness. In addition, this paper has conducted multitudinous simulation experiments on the GA‐based FSGA method and the traditional LSTM data analysis method respectively. The accuracy rate and other indicators are obtained by comparing the training. The experimental results show that the GA‐based FSGA machine learning method has higher prediction accuracy when analysing oversized economic data. And it is practical to accelerate the development of digital economy. [ABSTRACT FROM AUTHOR]

TE is a critical and difficult problem that tries to map traffic with various requirements to paths in dynamic communication networks. The emerging SDN enables centralized TE optimization with a global view. From the architectural viewpoint, ICN facilitates TE from many aspects, such as in-network caching which can reduce redundant traffic and content-awareness which can extract prior knowledge of content type directly. Thus, we leverage ICN to optimize SDN TE. However, ICN brings more complexities and dynamics to the network environment, which makes model-based TE methods inefficient. Inspired by recent advances in applying artificial intelligence techniques to solve complex online control problems, we investigate deep learning for content-awareness and DRL for TE decision. In addition, we propose a parallel online learning mechanism to safely utilize DRL which has trial-and-error nature. Results show that our proposal significantly improves network performance in terms of total network throughput, bandwidth utilization, and load balance. [ABSTRACT FROM AUTHOR]

Total antioxidant capacity (TAC) has become an important index to evaluate the food quality. Effective antioxidant detection has been the research hotspot of scientists. In this work, a novel three-channel colorimetric sensor array founded on Au 2 Pt bimetallic nanozymes for the discrimination of antioxidants in food was constructed. Benefiting from the unique bimetallic doping structure, Au 2 Pt nanospheres exhibited the excellent peroxidase-like activity with K m of 0.044 mM and V max of 19.37 × 10−8 M s−1 toward TMB. The density functional theory (DFT) calculation revealed that Pt atom in the doping system was active sites and there was no energy barrier in catalytic reaction which made Au 2 Pt nanospheres had excellent catalytic activity. Accordingly, a multifunctional colorimetric sensor array was constructed based on Au 2 Pt bimetallic nanozymes for rapid and sensitive detection of five antioxidants. Based on the different reduction ability of antioxidants, oxidized TMB could be reduced in different degrees. In the presence of H 2 O 2 , the colorimetric sensor array could generate differential colorimetric signals (fingerprints) by using TMB as the chromogenic substrate, which could be accurately discriminated through linear discriminant analysis (LDA) with a detection limit of <0.2 μM. The sensor array was able to the evaluate TAC in three actual samples (milk, green tea and orange juice). Furthermore, we prepared a rapid detection strip to meet the needs of practical application, making a positive contribution to food quality evaluation. • Au 2 Pt bimetallic nanozymes exhibited excellent peroxidase-like activity. • A nanozyme colorimetric sensor for detection of five antioxidants was constructed. • Pt atoms in the bimetallic nanozyme were active sites. • The evaluation of TAC in food was achieved by this bimetallic nanozyme sensor. • A rapid detection strip based on the bimetallic nanozyme was prepared. [ABSTRACT FROM AUTHOR]

Ontology is the kernel technique of Semantic Web (SW), which enables the interaction and cooperation among different intelligent applications. However, with the rapid development of ontologies, their heterogeneity issue becomes more and more serious, which hampers communications among those intelligent systems built upon them. Finding the heterogeneous entities between two ontologies, i.e., ontology matching, is an effective method of solving ontology heterogeneity problems. When matching two ontologies, it is critical to construct the entity pair's similarity feature by comprehensively taking into consideration various similarity features, so that the identical entities can be distinguished. Due to the ability of learning complex calculating model, recently, Artificial Neural Network (ANN) is a popular method of constructing similarity features for matching ontologies. The existing ANNs construct the similarity feature in a single perspective, which could not ensure its effectiveness under diverse heterogeneous contexts. To construct an accurate similarity feature for each entity pair, in this work, we propose an adaptive aggregating method of combining different ANNs. In particular, we first propose a context-based ANN and syntax-based ANN to respectively construct two similarity feature matrices, which are then adaptively integrated to obtain a final similarity feature matrix through the Ordered Weighted Averaging (OWA) and Analytic hierarchy process (AHP). Ontology Alignment Evaluation Initiative (OAEI)'s benchmark and anatomy track are used to verify the effectiveness of our method. The experimental results show that our approach's results are better than single ANN-based ontology matching techniques and state-of-the-art ontology matching techniques. [ABSTRACT FROM AUTHOR]

Ontology is the kernel technique of the Semantic Web (SW), which models the domain knowledge in a formal and machine-understandable way. To ensure different ontologies' communications, the cutting-edge technology is to determine the heterogeneous entity mappings through the ontology matching process. During this procedure, it is of utmost importance to integrate different similarity measures to distinguish heterogeneous entity correspondence. The way to find the most appropriate aggregating weights to enhance the ontology alignment's quality is called ontology meta-matching problem, and recently, Evolutionary Algorithm (EA) has become a great methodology of addressing it. Classic EA-based meta-matching technique evaluates each individual through traversing the reference alignment, which increases the computational complexity and the algorithm's running time. For overcoming this drawback, an Interpolation Model assisted EA (EA-IM) is proposed, which introduces the IM to predict the fitness value of each newly generated individual. In particular, we first divide the feasible region into several uniform sub-regions using lattice design method, and then precisely evaluate the Interpolating Individuals (INIDs). On this basis, an IM is constructed for each new individual to forecast its fitness value, with the help of its neighborhood. For testing EA-IM's performance, we use the Ontology Alignment Evaluation Initiative (OAEI) Benchmark in the experiment and the final results show that EA-IM is capable of improving EA's searching efficiency without sacrificing the solution's quality, and the alignment's f-measure values of EA-IM are better than OAEI's participants. [ABSTRACT FROM AUTHOR]

As the deployment of Internet of Things (IoT) devices becomes more widespread, the demand for data analysis services is also increasing in both range and volume. Despite the increasing demand for services due to the rise of IoT devices, efficiently processing latency-sensitive data analysis tasks (DATs) remains a significant challenge, especially in areas where communication infrastructure is limited or unavailable. Unmanned aerial vehicles (UAVs) can enhance communication coverage and quality by leveraging their flexible deployment characteristics, particularly in areas where network coverage is limited for IoT devices. UAVs can provide edge computing services while simultaneously offering network coverage, making them a valuable asset for improving communication in such areas. In this paper, we propose an Mobile Edge Computing (MEC) system that utilizes multiple UAVs equipped with edge servers to offer latency-sensitive analytical services to IoT devices on the ground. However, traditional task scheduling algorithms are challenging to adapt to dynamic and complex edge network environments, particularly for scheduling tasks with dependencies. To address this challenge, we employ deep reinforcement learning (DRL) to develop Directed Acyclic Graph (DAG) task scheduling algorithms and UAV deployment optimization algorithms. Our algorithm enables us to obtain optimal scheduling and deployment adjustment strategies in dynamic and changing environments. Simulation experiments demonstrate that in a MEC system comprising of multiple UAVs, our algorithm can swiftly converge to the optimal value, resulting in a significant reduction in DAT response time and cluster energy consumption compared to baseline algorithms. [ABSTRACT FROM AUTHOR]

The construction of Smart Cities is inseparable from the healthy operation of markets. Reasonable data analysis can provide a crucial foundation for the development of market behavior by considering the enormous amount of data generated by a market economy. To this end, we propose enhanced cluster generative adversarial networks (eClusterGAN) to achieve latent space clustering. However, data storage security is crucial. Moreover, we suggest a GAN-based network intrusion detection system (GAN NIDS) that uses adversarial learning to assist the generator in learning the spatial distribution of normal network flows. The simulation results showed that the proposed eClusterGAN and GAN NIDS outperformed the benchmarks in terms of clustering accuracy, running time, precision, recall, and F1, which can support researchers in studying economic data trends. The construction of Smart Cities can effectively ensure healthy market development by discovering and disseminating the potential value of market economic data. [ABSTRACT FROM AUTHOR]

• We propose a personalized course recommendation method based on BERT model, which integrates deep learning and big data technology based on MOOC system. • We design a domain feature difference learning strategy, which is able to learn the difference features between course contents to better extract information in the text and improve the recommendation performance of the model. • We experimentally demonstrate the effectiveness of the method proposed in this paper on the open MoocCube datasets. Finding the courses that users are interested in quickly in the massive data can make a very important contribution to the accurate dissemination of knowledge. In this paper, we integrate the deep learning and big data technology to investigate a personalized recommendation method based on Massive Open Online Course (MOOC) system. Based on the Bidirectional Encoder Representations from Transformers (BERT) model, we propose some corresponding strategies to improve the accuracy of the recommendation system. First, we introduce the acquisition and preprocessing of the open dataset. Second, we design a recommendation model framework by taking advantage of the BERT model and incorporating a self-attention mechanism. Finally, to obtain deep feature information between course texts, we design a domain feature difference learning strategy to improve the model's recommendation performance. The results of our experiments prove that the proposed model in this paper performs good recommendation results compared with other methods. [Display omitted] [ABSTRACT FROM AUTHOR]