With the prevalence of Internet AI technology, financial fraud becomes an imperative problem, especially in the context of machine learning. Technologies such as deep learning and natural language processing provide effective tools for detecting fraud with the guidance of financial statements, improving the efficiency and accuracy of data analysis, and helping to ensure financial safety. In this study, we propose a sophisticated representation learning method to detect financial statement fraud by tracking the detailed changes in the company’s Management Discussion and Analysis (MD&A) documents over time. Unlike traditional word frequency methods, we align paragraphs between consecutive disclosures based on their similarity at the paragraph level and categorize them into three types: added, deleted, and matched. Next, we create multivariate change trajectory representations based on fraud-related word categories. Finally, we use these word-level change trajectories to design a fraud detection model and compare it with several traditional models as well as the latest Time-Series Foundation Models. Experiments on 24 years of financial report data, from 1995 to 2019, show that our representation learning method significantly improves the performance of financial statement fraud detection across 11 different machine learning models, consistently outperforming traditional word frequency methods. Our method opens a new paradigm for feature engineering in financial statement fraud detection. Our code can be found at https://github.com/LittelStudent/Financial-Statement-Fraud-Detection-ParaEmb-FraudW2V .

The rapid dissemination of information through digital platforms has revolutionized the way we access and consume data, creating conditions that may lead to an increase in financial statement fraud, which jeopardizes the efficient functioning of capital markets. This paper propose a sophisticated representation learning method to detect financial statement fraud by tracking detailed changes in a firm’s Management Discussion and Analysis (MD&A) documents over time. Unlike traditional word frequency approaches, we start by aligning paragraphs between consecutive disclosures based on their representation-level similarities. Given the paragraph-embedding similarity, we categorize paragraphs into three types: added, deleted and matched. Next, we construct multivariate change trajectory representations based on fraud-relevant word categories, such as sentiment and uncertainties. Finally, we develop a fraud detection model using these word-level change trajectory representations. Extensive experiments on 24 years of financial report data, from 1995 to 2019, show that our representation learning approach significantly improves financial statement fraud detection performance across 7 different backbone machine learning models, consistently outperforming traditional word frequency-based approaches. Our method marks a new paradigm in feature engineering for financial statement fraud detection.

Quantum communications using entangled photons as qubits offer a promising approach to quantum key distribution, enhancing network security. However, establishing successful entanglement links between distant nodes is challenging due to factors like distance and rapid entanglement decay, resulting in low network transmission efficiency for quantum key distribution. This paper studies the problem of finding the maximum quantum communication network transmission scheme for multiple source-destination pairs with considering the success rate of quantum entangle establishments. We propose the Feasible Quantum Communication Network Transmission (fQUIET) algorithm, which leverages network transformation and auxiliary network construction. The Maximum Quantum commUnIcation nEtwork Transmission (mQUIET) algorithm is further proposed by iteratively searching for augmenting paths. Through theoretical analysis, we demonstrate that the mQUIET algorithm is capable of achieving the maximum QCN transmission scheme. Extensive experiments are carried out and the results reveal that our suggested algorithms have superior performance compared to existing methods.

The open resolver, which provides resolution services for the public without verifying user identities, is a critical component of the Domain Name System (DNS). In reality, users have natural trust in public resolution services, which is unilateral and unreliable due to the existence of open resolvers which tamper with the user’s DNS requests to the wrong IP addresses on the Internet. This tampering behavior has strong concealment which is difficult to be detected by users or security researchers. Therefore, it is exactly essential to discover such malicious behavior in time when combating cyber attackers. The traditional solution is to use the IP addresses of malicious open resolvers for blacklist defense, which is simple to implement but has disadvantages of false positives and false negatives. In this paper, we propose a method based on deep learning, namely Byte-level Autoencoder, to implement the detection of malicious open resolvers without the complex manual feature extraction process. Our work is the first study to detect malicious open resolvers in this field, based on the similarity of normal DNS messages. Experimental results show that our method has a high accuracy of 98.83%, a high precision of 97.84% and a low false positive rate of 0.11%, which verifies the effectiveness of our method.

Object detection methods for perspective images have proven increasingly efficient, but the techniques for equirectangular projection (ERP) panoramas from inherently spherical imaging cannot still achieve satisfactory performance. Due to the various degrees of distortion at different pixel locations, current algorithms cannot adapt to the changes in shape and contour caused by stretching, which results in performance degradation when migrating them from perspective images to spherical ones. In this paper, we improve the network for panorama object detection and introduce the cube-domain information with discontinuity but low distortion to correct the panorama features. Unlike previous works, we consider the impact of semantic discontinuity from all tangent planes instead of overlaying features when needed. Considering the six facets as unified, i.e., six-to-one for extraction, the proposed Facet-Link module enhances the long-range sensing capability at the facet level in the frequency domain. Moreover, the position alignment packs different facets, i.e., six-to-one for calibration, to preserve more global signals during the correction stage, which establishes semantic pathways for feature interactions between panorama and cubemap in the two dimensions, facet-facet and cube-pano, respectively. Extensive experiments on synthetic and real-world datasets verify the effectiveness and robustness of our proposed method.

Abstract Hunting the advanced threats hidden in the enterprise networks has always been a complex and difficult task. Due to the variety of attacking means, it is difficult for traditional security systems to detect threats. Most existing methods analyze log records, but the amount of log records generated every day is very large. How to find the information related to the attack events quickly and effectively from massive data streams is an important problem. Considering that the knowledge graph can be used for automatic relation calculation and complex relation analysis, and can get relatively fast feedback, our work proposes to construct the knowledge graph based on kernel audit records, which fully considers the global correlation among entities observed in audit logs. We design the construction and application process of knowledge graph, which can be applied to actual threat hunting activities. Then we explore different ways to use the constructed knowledge graph for hunting actual threats in detail. Finally, we implement a LAN-wide hunting system which is convenient and flexible for security analysts. Evaluations based on the adversarial engagement designed by DARPA prove that our platform can effectively hunt sophisticated threats, quickly restore the attack path or assess the impact of attack.

With the wide application of HTTPs, malware HTTPs traffic classification is usually the first step in anomaly detection system. The existing classification methods mainly use the raw bytes (containing the discriminative features) or the statistical features (containing the global information) as the input, which leads to a low Fl-score. Therefore, this paper presents a novel Attention-based Fusion Network (FA-net), which combines two types of features properly to improve the classification performance. FA-net consists of three sub-networks: RF -net and SF -net extract the representative features of raw bytes and statistical features through the Convolutional Neural Network (CNN) and reconstruction mechanism respectively, and C-net combines two types of features through the attention mechanism and a regulating factor. The experiments indicate that FA-net obtains markedly better results (the average Fl-score of 0.941 and 0.997 respectively on two datasets) than the baselines. We also explore the influence of different regulating factor values on classification performance.

As one type of omnidirectional projection, fisheye images have been widely used in automatic driving and visual surveillance. However, they cannot be processed well by the traditional algorithms designed for the planar rectilinear images since they usually suffer from severe geometric distortion during image formation. In this paper, the conventional face detection algorithm is enhanced to fit the fisheye images via combining with the spherical convolution block by learning rotation-invariant features from the spherical domain. The learned features from both planar and spherical domains are subsequently mixed by the spatial attention mechanism. Consequently, the whole network can automatically learn the distorted features directly from different positions on the target image. Experimental results verify that our network can detect distorted faces on fisheye images effectively and maintain the performance on traditional planar images.

The key challenge for large scale software system maintenance is to minimize the troubleshooting time when severe system anomaly (e.g., server failure, link congestion, software bugs) happens. It often takes hours for operators to manually locate the fault and thus degrades the service performance in terms of user experience and economics. Previous root cause localization algorithms are usually time-consuming and error-prone. In this paper, we present AutoRoot, a fast and accurate multi-dimensional root cause localization algorithm. Specifically, AutoRoot uses an adaptive density clustering to improve the accuracy and an effective filtering mechanism to reduce the search time. Extensive experiments using multiple real data traces validate the performance of AutoRoot compared with existing algorithms.

With the proliferation of edge intelligence and the breakthroughs in machine learning, Federated Learning (FL) is capable of learning a shared model across several edge devices by preserving their private data from being exposed to external adversaries. However, the distributed architecture of FL naturally introduces communication between the central parameter server and the distributed learning nodes. The huge communication cost poses a challenge to practical FL, especially for FL in mobile edge computing (MEC) networks. Existing communication-efficient FL systems predominantly optimize their intrinsic learning process and are not concerned with the implications on the network. In this paper we propose a FL scheme that leverages Device-to-Device (D2D) communication (hence called D2D-FedAvg) and is suitable for mobile edge networks. D2D-FedAvg creates a two-tier learning model where D2D learning groups communicate their results as a single entity to the MEC server leading to traffic reduction. We propose the schemes for D2D grouping, master UE selection, and also D2D exit in the learning process and then form a complete D2D-assisted federated averaging algorithm. Via extensive simulations on the Federated Extended MNIST dataset, the feasibility and convergence of D2D-FedAvg scheme are evaluated. Our results show that D2D-FedAvg lowers the communication cost relative to the typical Federated Averaging (FedAvg) in cellular networks as the number of users is increased (for 100 cellular users 37% traffic reduction), while keeping the same learning accuracy with FedAvg across the board.

Abstract Long-term prediction is still a difficult problem in data mining. People usually use various kinds of methods of Recurrent Neural Network to predict. However, with the increase of the prediction step, the accuracy of prediction decreases rapidly. In order to improve the accuracy of long-term prediction,we propose a framework Variational Auto-Encoder Conditional Generative Adversarial Network(VAECGAN). Our model is divided into three parts. The first part is the encoder net, which can encode the exogenous sequence into latent space vectors and fully save the information carried by the exogenous sequence. The second part is the generator net which is responsible for generating prediction data. In the third part, the discriminator net is used to classify and feedback, adjust data generation and improve prediction accuracy. Finally, extensive empirical studies tested with five real-world datasets (NASDAQ, SML, Energy, EEG,KDDCUP)demonstrate the effectiveness and robustness of our proposed approach.

In the past decades, various approaches have been proposed to address the time series prediction problem, among which nonlinear autoregressive exogenous (NARX) models achieve great progresses in one-step time prediction. Although NARX models are capable of capturing long-term dependence of the time series data, the impact of associated attributes lacks enough attention. To cope with this issue, in this paper we propose a Multi-Attention algorithm based Recurrent Neural Network (RNN) to perform multivariate time series forecasting. In the first stage, given a raw multivariate time series segment, we obtain both relevant encoder hidden state and encoder hidden state of the associated attribute by employing input-attention and self-attention respectively. In the second stage, we use temporal-convolution-attention neural network to process the encoder hidden states and capture long-range temporal patterns. Finally, extensive empirical studies tested with four real world datasets (NASDAQ100, SML2010, Gas Sensor Array Temperature Modulation and Air Quality) demonstrate the effectiveness and robustness of our proposed approach.

Software defined networking is attracting extensive attention to achieve better network performance through flexible flow path control. However, traditional OpenFlow-based forwarding method faces practical challenges such as limited flow table size and dynamic flow path setup. In this work we propose a scalable and flexible path control schema based on source routing to address those problems through pre-installing flow paths into the data plane and inserting path information into the packet header at the network edge. To reduce the flow table overhead, we assign an ID to each path and aggregate IDs by wildcards. We formulate the path ID assignment problem and decompose it into the link ID assignment problem, which is solved by our proposed heuristic algorithm based on vertex coloring and topological potential. Extensive simulations on various topologies show that our proposed algorithm can reach the optimal solutions in most topologies within the time limit, and our path control schema can save more than 98% rules compared with traditional OpenFlow-SDN method and has good universality. In addition, our method can achieve better performance in terms of dynamic flow path setup due to the characteristic of source routing.

In the context of Advanced Persistent Threat-s (APTs), system audit log-based intrusion forensics has been proposed to carry out attack investigation. System audit log is highly suitable for intrusion forensics because it records the interactions among system entities in detail. However, system audit log has a fatal shortcoming due to its massive growth of log size. To address this issue, this paper proposes a compression scheme for system audit log named T-Tracker. Firstly, T-Tracker detects the events that communicate with external data sources and generates the initial taint set. Then it tracks the diffusion of the taint according to the audit log. By retaining the events on diffusion path only, we can achieve log compression. Our evaluation with different system workloads and attack cases demonstrates that our approach can achieve significant log compression without affecting the accuracy of intrusion forensics.

We proposed an intelligent website classification schema based on deep neural networks using mixed featured extractors. With the guidance of supervised learning methods and iterative training, we use the gradient descent algorithm to model the website classification. This novel model is composed of four components, which includes a Website Encoder, a Text CNN Feature Extractor, a Bidirectional GRU Feature Extractor and a Fully Connected Classifier. It can extract multiple features at different granularities of a website. By using the concatenated mixed features taken from mixed feature extractors, our model can easily choose a suitable website class. We make extensive experiments on the realistic collected website dataset. The dataset is collected using domains extracted from DNS records of Telecom Operator. Compared the multiple widely used machine learning models and our novel model, results demonstrate the proposed classification schema outperforms the current models with the metrics precision, recall, F1, and accuracy. All of this can contribute to various web applications, such as malicious website detection, online advertising, etc.

Multi-path TCP has recently shown great potential to take advantage of the rich path diversity in data center networks (DCN) to increase transmission throughput. However, the small flows, which take a large fraction of data center traffic, will easily get a timeout when split onto multiple paths. Moreover, the dynamic congestions and node failures in DCN will exacerbate the reorder problem of parallel multi-path transmissions for large flows. In this paper, we propose DC2-MTCP (Data Center Coded Multi-path TCP), which employs a fast and light-weight coding method to address the above challenges while maintaining the benefit of parallel multi-path transmissions. To meet the high flow performance in DCN, we insert a very low ratio of coded packets with a careful selection of the packets to be coded. We further present a progressive decoding algorithm to decode the packets online with a low time complexity. Extensive ns2-based simulations show that with two orders of magnitude lower coding delay, DC2-MTCP can reduce on average 40% flow completion time for small flows and increase 30% flow throughput for large flows compared to the peer schemes in varying network conditions.

Cloud computing has become a promising paradigm in the field of Services Computing for its advantage of providing infrastructures and resources as a service. However, cloud networks consume huge amount of energy consumption due to the various applications and large scale of users with high mobility. In this paper, we investigate the problem of service migration to minimize the energy consumption of cloud infrastructures while ensuring the service delivery with satisfied latency and economic cost. We propose an energy-efficient migration scheme named SMPA considering the economic aspects and the quality of service. Furthermore, the mechanism of virtual machine consolidation running different services is also designed in SMPA, when the requests arrive, change or depart. Extensive simulations conducted on random topologies show that our proposal reduces requests' SLA violation ratio and improves the utilization of provider resources by service migration. Moreover, by taking advantage of physical resources, SMPA decreases the number of running hosts and reduces the power consumption while ensuring the operational cost effectively under dynamic workload scenarios.

With the increasing popularity of cloud computing, the huge amount of energy consumed by data centers has gained much attention. Current proposals address the energy efficiency problem by two major methods: optimizing the allocation of physical servers and network elements (routers or switches). In order to improve resource utilization and minimize the energy consumption, the former method focuses on virtual machine (VM) placement regardless of the inherent traffic between VMs. The latter designs power saving routing and flow scheduling while this method neglects the resource demands in VMs. In this paper, we jointly consider the VM placement and network routing to optimize energy cost. In addition, we take advantage of the centralized and global controller in Software Defined Networking (SDN) paradigm. Inspired by the idea of Data Field, we propose a novel algorithm to evaluate the importance and relationships among multiple VMs. The potential score based on Data Field is a more accurate global ranking considering the resource demands and inter-VM traffic. Extensive simulations are conducted on different scales of typical data center topologies, such as Fat-Tree and BCube. Results show that our proposal can reduce the number of active devices including the servers and network elements, and thereby saves power consumption. Moreover, the proposed algorithm improves network performance by decreasing the average hops of per flow.

In this paper, we study the VNF placement problem in hybrid NFV environment, which is important during the transition from traditional networks to NFV networks. We first propose a new concept of hybrid NFV environment, which is more comprehensive and realistic than the former works. Then, we give out a novel model of VNF placement optimization to achieve lower bandwidth consumption and lower maximum link utilization simultaneously, with consideration of VNF combination. Next, to solve this problem, we propose four genetic algorithms, which are combinations of the frameworks of two existing algorithms (MOGA and NSGA-II) and our novel modifications. Simulation results show that, in our 4 algorithms Greedy-NSGA-II has the best performance. When compared with other two non-genetic algorithms (BM and Random), the average total bandwidth consumption of Greedy-NSGA-II is only 12.24% and 2.96% of theirs respectively, and the average maximum link utilization of Greedy-NSGA-II is only 25.04% and 13.81% of theirs respectively.

It is a challenge to improve utilization of physical resources when allocating resources for virtual data centers (VDCs). In this work we proposed two reliability-aware VDC embedding methods NMP and NMPD. To reduce embedding cost, the proposals try to split VMs into clusters based on topological potential and modularity and assign each cluster to a server. Via simulations, we show that NMP can accept more VDCs and help InPs improve the revenue. Although accepting a little less VDCs, NMPD can reduce resources occupied by VDCs and alleviate the congestion of core links.

Abstract In wireless sensor networks, one of the primary requirements is that sensor data acquired from the physical world can be interchanged with all interested collaborative entities in a secure, reliable manner. Because of highly unpredictable nature of the environments caused by malicious attacks or potential threats, minimizing transmission cost between source and sink nodes with jointly considering the security of the whole network is a critical issue. This paper considers two optimization problems of deriving the minimum cost paths from multiple source nodes to the sink node under the guaranteed level of the vulnerability. The link or node vulnerability is defined as a metric, which characterizes the degree of link or node sharing among paths. With the defined link vulnerability, the link vulnerability‐constrained minimum cost paths problem is first formulated, and two polynomial‐time algorithms are developed for deriving the optimal paths. For the node‐vulnerability‐constrained minimum cost paths problem, we adopt the network conversion and then achieve the optimal solution with previous proposed algorithms. The necessary condition for solution existence, the optimality of the proposed algorithms, and the related properties of tree network are further theoretically analyzed. Extensive simulations show the significant performance improvements achieved by our proposed algorithms.Copyright © 2014 John Wiley & Sons, Ltd.

Abstract Recently, benefiting from rapid development of energy harvesting technologies, the research trend of wireless sensor networks has shifted from the battery‐powered network to the one that can harvest energy from ambient environments. In such networks, a proper use of harvested energy poses plenty of challenges caused by numerous influence factors and complex application environments. Although numerous works have been based on the energy status of sensor nodes, no work refers to the issue of minimizing the overall data transmission cost by adjusting transmission power of nodes in energy‐harvesting wireless sensor networks. In this paper, we consider the optimization problem of deriving the energy‐neutral minimum cost paths between the source nodes and the sink node. By introducing the concept of energy‐neutral operation, we first propose a polynomial‐time optimal algorithm for finding the optimal path from a single source to the sink by adjusting the transmission powers. Based on the work earlier, another polynomial‐time algorithm is further proposed for finding the approximated optimal paths from multiple sources to the sink node. Also, we analyze the network capacity and present a near‐optimal algorithm based on the Ford–Fulkerson algorithm for approaching the maximum flow in the given network. We have validated our algorithms by various numerical results in terms of path capacity, least energy of nodes, energy ratio, and path cost. Simulation results show that the proposed algorithms achieve significant performance enhancements over existing schemes. Copyright © 2016 John Wiley & Sons, Ltd.

Cloud computing is a promising paradigm for future computing platform. It enables the physical resources(computing, storage, networking, etc) to be provided on-demand. Despite the cloud computing brings many benefits to the IT network structure, it still faces new challenges for allocating heterogeneous resources automatically to services. Software Defined Networking (SDN) is a new concept of the network infrastructure as it decouples the control and data planes.With its characteristic of programmability, it's feasible to achieve network virtualization or create a slice of network in the form of virtual data centers(VDCs), which have pushed huge development of cloud computing. The VDCs embedding problem deals with allocating VDC requests to fulfill the requirements of cloud services with minimal cost and high revenue, which is an NPhard problem. In this paper, we propose an algorithm based on Entropy Weighted topological Potential considering multiple types of resources to tackle the problem in SDN-enabled data center network. Extensive simulations show that our proposed algorithm can embed more virtual requests efficiently with higher acceptance ratio and resource revenue over time.

TCP Incast refers to the phenomenon of goodput collapse when multiple synchronized servers send data to the same client in parallel. In this paper, we propose a novel application-level scheduling approach named OSDT (Optimal Staggering Data Transfers) for TCP Incast problem. OSDT limits the number of concurrent TCP flows as well as servers' sending rate to optimal values so that the utilization of link capacity can be maximized without any packet losses. To achieve this, we build an optimization model with the usage of network and application information. Based on this model we can get the optimal values for the parameters in OSDT. Simulation results indicate that OSDT can achieve the highest goodput among all existing application-level scheduling approaches in a wide range of network and application parameters, and its performance is stable. So OSDT can be seen as an effective and scalable solution for TCP Incast problem.

Due to the special topologies and communication pattern, in today's data center networks it is common that a large set of TCP flows and a small set of TCP flows get into different ingress ports of a switch and compete for a same egress port. However, in this case the throughput share of flows in the two sets will not be fair even though all flows have the same RTT. In this paper, we study this problem and find that TCP's fairness in data center networks is related with not only the network capacity but also the number of flows in the two sets. We propose a mathematical model of the average throughput ratio of the large set of flows to the small set of flows. This model can reveal the variation of TCP's fairness along with the change of network parameters (including buffer size, bandwidth, and propagation delay) as well as the number of flows in the two sets. We validate our model by comparing its numerical results with simulation results, finding that they match well.

Recently, energy harvesting technology has been introduced into wireless sensor networks to solve the traditional battery-powered energy bottleneck problem. However, due to battery capacity limitation, the harvested energy would overflow while the nodes are in the energy saturation status. Aiming at this, we consider topology control approach in the EHWSNs that allows each node to adaptively adjust its transmission power level to utilize the harvested energy efficiently. Specifically, we first model nodes' behaviors as an ordinal potential game where the high harvesting power nodes interact with the low harvesting power nodes to collaboratively maintain the whole network's connectivity. And we theoretically prove the existence of Nash equilibrium in this game. Then, a polynomial-time algorithm has been proposed to achieve Nash equilibrium. Simulation results show that our algorithm exploits the available energy resources in an efficient way and outperforms existing energy-aware algorithm in terms of energy conservation and equilibrium distribution.

In wireless sensor networks (WSNs), sensor nodes are usually powered by batteries of limited capacity, which results in dynamic changes of available paths for data aggregation due to node failures caused by energy depletion. For transmitting certain amount of data generated by source node, the overall transmission cost is affected by two major factors, using sequence of available paths and amount of data imposed on each path, which becomes a major issue significantly influencing the efficient usage of the networks. To address this issue, we consider the optimization problem of how to minimize the overall transmission cost of given data delivered from the source node to the sink node in the energy-constrained WSN. Specifically, we first describe the problem on the basis of the minimum cost flow theory and derive the upper bound for the data amount in terms of the number of packets that can be successfully transmitted from the source node to the sink node. Then, we propose specific algorithms to derive the optimal paths and their optimal data amounts, and then achieve the minimized overall transmission cost for the certain amount of data. Extensive simulations show that significant performance enhancement can be achieved by using our proposed algorithms.

In wireless sensor networks, one of the primary requirements is that sensor data derived from the physical world can be interchanged with all interested collaborative entities in a secure and reliable manner. Due to highly unpredictable environments where sensor nodes are usually deployed, minimizing the transmisstion cost with jointly taking into account the security of the whole network poses a challenging task. With this end, this paper considers an optimization problem of deriving the minimum cost paths from multiple source nodes, which are deployed in the area of interest, to the sink node under the constraint that the vulnerability of the whole network is under the given level. The vulnerability is defined as a metric which characterizes the degree of edge and node sharing among different paths. With the defined vulnerability, vulnerability-constrained minimum cost paths problem is formulated and two polynomial-time algorithms are developed for deriving the optimal paths. The necessary condition for the existence of the optimal solution, and the optimality of the proposed algorithms are analyzed in the theoretical. Extensive simulations show the significant performance enhancements achieved by our proposed algorithms.

Network virtualization provides a powerful way to run multiple heterogeneous virtual networks (VNs) at the same time on a shared substrate network. A major challenge in network virtualization is the efficient virtual network embedding: mapping virtual nodes and virtual edges onto substrate networks. Previous researches have presented several heuristic algorithms, which fail to consider the topology attributes of substrate and virtual networks. However, the topology information affects the performance of the embedding obviously. In this paper, for the first time, we exploit the topology attributes of substrate and virtual networks, introduce network centrality analysis into the virtual network embedding, and propose virtual network embedding algorithms based on closeness centrality. Due to considering the topology information, our study is more reasonable than the existing work in coordinating node and edge embedding. In addition, with the guidance of topology quantitative evaluation, the proposed network embedding approaches largely improve the network utilization efficiency and decrease the embedding complexity. Experimental results demonstrate the usability and feasibility of the proposed approach.