Search

With the rapid development and integration of biomedicine and information technology,massive amounts of imaging data,patient report data,electronic health records,and omics data have been accumulated rapidly in healthcare.These data are cha-racterized by complexity,heterogeneity and high dimensionality.Deep learning has the ability of complex function simulation and automatic feature learning,which can provide efficient technical support for research in medical diagnosis and drug development.Currently,deep learning has been extremely successful in medical imaging and further more,some medical imaging diagnostic systems based on deep learning have achieved performance that is even comparable to that of relevant experts.Due to the progress of natural language processing technology,deep learning has also made remarkable progress in the use of non-image data tasks.This paper first briefly describes the development of deep learning in healthcare.Subsequently,the application of deep learning model in healthcare is statistically analyzed,and some available datasets are sorted out.In addition,this paper also introduces the research progress of deep learning in medical diagnosis and treatment processes such as disease diagnosis and health monitoring,and its research progress in protein structure prediction and drug discovery.Finally,key challenges of deep learning in healthcare applications such as data quality,interpretability,privacy security and practical application limitations are discussed.It also discusses feasible solutions or approaches to these challenges.

Graph is an important and fundamental data structure that presents in a wide variety of practical scenarios.With the rapid development of the Internet in recent years,there has been a huge increase in social network graph data,and the analysis of this data can be of great help in practical scenarios such as public services and advertising and marketing.There are already quite a few graph neural network algorithms that can get good results in such problems,but they still have room for improvement,and in many scenarios where high accuracy is pursued,engineers still want to have algorithms with better performance to choose from.This paper improves personalized propagation of neural predictions and proposes a new graph neural network algorithm called degree of node based personalized propagation of neural predictions(DPPNP)that can be used in social graph networks.Compared to traditional graph neural network algorithms,when the information is propagated between nodes,the proposed algorithm will keep its own information for different nodes in different proportions according to the degree of nodes,so as to improve the accuracy.Experiments on real datasets show that the proposed algorithm has better performance compared to previous graph neural network algorithms.

With the rapid development of social media and the popularity of mobile devices,the interaction between users has become more convenient.But at the same time,rumors on social media are more and more rampant,which brings hidden dangers to the public and social safety.In the real world,users often express their own opinions after observing other microblogs that have been posted,especially the context of the microblog to be replied.Although some existing rumor detection methods learn the propagation patterns on propagation trees of rumors to extract clues of user interrogation or factual evidences based on the principle of crowd wisdom,which greatly improves the performance of rumor detection,they only focus on those microblogs that have direct response relationships,and Lack of ability to fully mine the indirect and implicit relationships among microblogs in the process of rumor propagation.Therefore,in this paper,a node and path dual-attention network on propagation tree structures(DAN-Tree) for debunking rumors is proposed.First,the model uses the Transformer structure to fully learn the implicit semantic relationship between posts in the propagation path,and then uses the attention mechanism to perform weighted fusion to obtain the feature vector of the propagation path.Secondly,the path representation is weighted and aggregated by using the attention mechanism to obtain the representation vector of the whole propagation tree.In addition,the structure embedding method is used to learn the spatial location information of the post on the propagation tree,which realizes the effective fusion of the deep structure and semantic information in the rumor propagation structure.The effect of the DAN-Tree model is verified on four classic datasets.Experimental results show that the DAN-Tree model surpasses the best results of the existing literature on the three datasets:the accuracy of the Twitter15 and Twitter16 datasets increases by 1.81% and 2.39%,respectively,and the F1 score of the PHEME dataset increases by 7.51%,which proves the effectiveness of DAN-Tree model.

Nearest neighbor search has become a significant research problem due to its wide applications.Traditional spatial index structures such as R-tree and KD-tree can efficiently return accurate nearest neighbor search results in low-dimensional space,but they are not suitable for high-dimensional space.Locality sensitive B-tree(LSB) hashes data points to the sortable one-dimension values and arranges them in a tree-like structure,which dramatically improves the space and query efficiency of the previous locality sensitive hashing(LSH) implementations,without compromising the resulting quality.However,LSB fails to take data distribution into account.It performs well in a uniform data distribution setting,but exhibits unstable performance when the data are skewed.In response to this problem,this paper proposes LayerLSB,which reconstructs the hash values in a dense range by exploring the density of the hash values to make the distribution more uniform,so as to improve the query efficiency.Compared to LSB,LayerLSB indices become more targeted in terms of data distribution,and a multi-layered structure is constructed.Compared with the simple rehashing method,the multi-layered approach will still guarantee the search quality by carefully choosing the number of groups and hash functions.The results show that the query cost can be reduced to 44.6% of the original at most when achieving the same query accuracy.

Traffic flow forecasting is the cornerstone of many applications in transportation which has a great importance in smart city system.The difficulty of this task is how to effectively model the temporal and spatial dependence.Existing methods usually use GNN to model temporal correlation and CNN or RNN to model temporal correlation.When modeling the spatial correlation,only the adjacency matrix is applied to model local relationships while ignoring global spatial information.However,there are some roads in the entire road network whose surrounding structures are similar,and these roads carry similar functions in the road network.Therefore,the characteristics of these similar roads can also be used as the basis for traffic prediction.This paper proposes a traffic flow forecasting model based on multi-adjacent matrix and multi-head attention mechanism.It includes:1)the node2vec algorithm is applied to calculate the vector representation of the road in road network,and the similarity matrix is calculated through the threshold for graph convolution operation to extract global spatial information;2)the multi-channel self-attention mechanism is used to mine the spatial and temporal features of the model.Experiments on public datasets PEMS04 and PEMS08 demonstrate the proposed model’s effectiveness.Its accuracy is improved compared with the mainstream models.

Since 2018,scholars have proposed and studied a new topic of “concept reduction” in formal concept analysis.Including unnecessary concepts,core concepts,relatively necessary concepts,and the identification of three types of concepts,and research on concept reduction algorithm.In this paper,the same effect relation is proposed and its important properties are studied.It pre-sents a simple way to identify three types of concepts through the same effect relationship,and proposes a new algorithm for concept reduction which is based on the concept lattice of the complement set of subsets of the same effect relationship.For decades,the algorithm of “reduction topic” has adopted the method of conjunctive normal form and disjunctive normal form transformation.Many scholars even said that “reduction problem” was equivalent to the transformation of conjunctive paradigm and disjunctive paradigm.This paper studies a new method to solve the “reduction problem” without using the transformation between conjunctive normal form and disjunctive normal form.This new method is of significance both in theory and in practice.It is a new attempt.There are often many “concept reduction” in a background,so it is not very meaningful to find out all the results.Gene-rally,it is necessary to find out “concept reduction” containing some concepts,and the method in this paper has particular advantages in this respect.

Multiple object tracking(MOT)aims to predict trajectories of all targets and maintain their identities from a given video sequence.In recent years,MOT has gained significant attention and become a hot topic in the field of computer vision due to its huge potential in academic research and practical application.Benefiting from the advancement of object detection and re-identification,the current approaches mainly split the MOT task into three subtasks:object detection,re-identification feature extraction,and data association.This idea has achieved remarkable success.However,maintaining robust tracking still remains challenging due to the factors such as occlusion and similar object interference in the tracking process.To meet the requirement of accurate,robust and real-time tracking in complex scenarios,further research and improvement of MOT algorithms are needed.Some review literature on MOT algorithms has been published.However,the existing literatures do not summarize the tracking approaches comprehensively and lack the latest research achievements.In this paper,the principle of MOT is firstly introduced,as well as the challenges in the tracking process.Then,the latest research achievements are summarized and analyzed.According to the tracking paradigm used to complete the three subtasks,the various algorithms are divided into separate detection and embedding,joint detection and embedding,and joint detection and tracking.The main characteristics of various tracking approaches are described.Afterward,the existing mainstream models are compared and analyzed on MOT challenge datasets.Finally,the future research directions are prospected by discussing the advantages and disadvantages of the current algorithms and their development trends.

Adversarial reprogramming is an attacking method against the deep neural networks.By adding a certain perturbation to the input image,the network could be made to execute the attacker’s specified task,i.e.,destroying the legitimate permission of the training network model.It is positive to deeply understand and investigate this kind of attacks for further designing the corresponding anti-reprogramming algorithms.This paper discusses the relationship between the location of perturbations and the performance of adversarial reprogramming.Specifically,the noise visibility function is used to evaluate the adversarial distortion for each local content,and obtain the masking matrix.Then,the adversarial perturbations are added adaptively to optimize the attacking task.Experimental results show that,for the state-of-the-art deep network models,the proposed algorithm can enhance the performance of adversarial reprogramming attack and improve the imperceptibility of modified image.

Deep learning-based image compressive sensing has received extensive attention due to its powerful learning ability and fast processing speed.With the increase in the depth of convolutional neural networks,the existing image reconstruction methods using neural networks do not fully utilize the residual features in the network.In order to solve this problem,this paper proposes a compressed sensing attention neural network framework based on residual feature aggregation(RFA2CSNet)by jointly optimizing the sampling and inverse reconstruction processes.First,the block compressed sensing sampling sub-network and the initial reconstruction sub-network are constructed to adaptively learn the measurement matrix and generate the initial reconstruction image.Then the residual learning and spatial attention mechanisms are introduced to construct the residual feature aggregation attention reconstruction sub-network to make the residual feature more focused on the key spatial content,so as to further improve the reconstructed image quality.Experimental results show that the proposed network is superior to the existing image compressed sensing reconstruction algorithm in the case of comparable reconstruction time,and obtains better image compressive sensing reconstruction quality.Specifically,using 11 images for testing with a sampling rate of 0.10,the average peak signal-to-noise ratio increases by 0.34~6.18 dB compared with other deep learning-based methods.

Aiming at the problem of lacking of interpretability of deep image denoising networks,this paper proposes a multi-scale convolutional sparse coding network(MSCSC-Net)for image denoising using the idea of deep unfolding.Firstly,a multi-scale convolutional sparse coding(MSCSC)model is developed by exploiting the multi-scale convolutional dictionary,which can effectively express the multi-scale structure of image.Then,the traditional iterative optimization solution for solving the MSCSC model is unfolded into a deep neural network,namely MSCSC-Net.Each layer of MSCSC-Net corresponds to each iteration of the optimization solution.Therefore,the parameters of MSCSC-Net can be accurately defined through the traditional optimization model,which improves the interpretability.In addition,in order to preserve the structural of original image,the proposed MSCSC-Net adopts a revised residual learning idea,in which the weighted results of input noisy image and intermediate denoised image of previous layer are used as the input of next layer.Such revised residual learning can improve denoising performance further.Experimental results on public datasets show that MSCSC-Net is competitive to existing typical deep learning-based algorithms.Speci-fically,for the CBSD68 dataset at noise level 75,MSCSC-Net obtains 0.77% and 2.2% improvements over FFDNet in terms of the average PSNR and SSIM,respectively.

With the growth of Internet multimedia data,text image retrieval has become a research hotspot.In image and text retrieval,the mutual attention mechanism is used to achieve better image-text matching results by interacting image and text features.However,this method cannot obtain image features and text features separately,and requires interaction of image and text features in the later stage of large-scale retrieval,which consumes a lot of time and is not able to achieve fast retrieval and ma-tching.However,the cross-modal image text feature learning based on Transformer has achieved good results and has received more and more attention from researchers.This paper designs a novel Transformer-based text image retrieval network structure(HAS-Net),which mainly has the following improvements:a hierarchical Transformer coding structure is designed to better utilize the underlying grammatical information and high-level semantic information;the traditional global feature aggregation method is improved,and the self-attention mechanism is used to design a new feature aggregation method;by sharing the Transformer coding layer,image features and text features are mapped to a common feature coding space.Finally,experiments are conducted on the MS-COCO and Flickr30k datasets,the cross-modal retrieval performance has been improved,and it is in a leading position among similar algorithms.It is proved that the designed network structure is effective.

Due to its rich and intuitive expressivity,knowledge graph has received much attention of many scholars. A lot of works have been accumulated in knowledge graph embedding. The results of the works have played an important role in some fields, such as e-commerce, finance,medicine, transportation and intelligent Q & A. In the knowledge graph embedding model, the loss function plays a key role in its training stage. Based on the existing research of knowledge graph embeddings, this paper classifies the loss functions used in the model into six categories: hinge loss, logistic loss, cross entropy loss, log likelihood loss, negative sampling loss and mean square error loss. The prototype formula and physical meaning of loss functions and their expansion, evolution and application in knowledge graph embedding models are analyzed in detail one by one.Based on the above,the usage, efficiency and convergence of various loss functions in the static and dynamic knowledge graph scenarios are comprehensively analyzed and evaluated. According to the analysis results, combined with the development and application trend of knowledge graph and the current situation of loss functions,the future works of loss functions are discussed.

With the rapid growth of urban populations,the number of private cars has grown exponentially,which makes overwhelming traffic congestion problem become more and more acute.The traditional traffic signal control technology is difficult to adapt to the complex and changeable traffic conditions,and the data-driven methods bring new research directions for the control-based system.The combination of deep reinforcement learning and traffic control systems plays an important role in adaptive traffic signal control.First,this paper reviews the latest progress in the application of intelligent traffic signal control systems,the methods of intelligent traffic signal control are classified and discussed,and the existing works in this field are summarized.The deep reinforcement learning method can effectively solve the problems of inaccurate state information acquisition,poor algorithm robust and weak regional coordination control ability in intelligent traffic signal control.Then,on the basis of the above,this paper gives an overview of the simulation platforms and experimental setup for intelligent traffic signal control,and analyzes and verifies it through examples.Finally,The challenges and unsolved problems in this field are discussed and future research directions are summarized.

Keyphrase is a set of phrases that summarizes the core theme and key content of a given text.At present,information overload is becoming more and more serious,it is crucial to predict phrases with their central ideas for a given large amount of textual information.Therefore,keyphrase prediction,as one of the basic tasks of natural language processing,has received more and more attention from research scholars.Its corresponding methods mainly contain two categories,namely keyphrase extraction and keyphrase generation.Keyphrase extraction is the fast and accurate extraction of salient phrases that appear in the given text.Unlike keyphrase extraction,keyphrase generation predicts both phrases that appear in the given text and those do not appear in the given text.In summary,both have their advantages and disadvantages.However,most of the existing work on keyphrase ge-neration has ignored the potential benefits that extractive features may bring to keyphrase generation models.Extractive features can indicate important fragments of the original text and play an important role for the model to learn the deep semantic representation of the original text.Therefore,combining the advantages of extractive and generative approaches,this paper proposes a new keyphrase generation model incorporating multi-granularity extractive features(MGE-Net).Compared with recent keyphrase ge-neration models on a series of publicly available datasets,the proposed model achieves significant performance improvements in most evaluation metrics.

As a new network architecture,the software defined network(SDN)simplifies the network management logic by separating data plane and control plane,which is one of the popular research subjects of next-generation network.However,due to frequent link failures and other factors,it is difficult to guarantee the reliability of SDN,which is a problem well recognized in the industry.The existing SDN link failure recovery models often have the problems of long recovery delay,requiring too many flow entries and ignoring link congestion after recovery from failure.To solve these problems,this paper proposes a single-link failure recovery model(LFA-CA)based on segment routing(SR).The model employs the two heuristic algorithms of BPF and BPU to calculate a loop-free backup path during network initialization and update the congestion avoiding backup path during operation,respectively,so as to achieve fast recovery from single-link failure and congestion avoidance after failure.In this paper,massive simulation experiments are carried out to evaluate the performance of our model,and the results prove that compared with some of the existing SDN single-link failure recovery models,LFA-CA consumes less forwarding rules and has better load balancing ability after failure.

In optic fiber communication,traditional optical performance monitoring(OPM) mainly relies on analyzing the time-frequency domain information of the signal.However,conventional methods cannot complete multi-task joint monitoring,so they are less flexible.With the development of machine learning,the monitoring of optical signal modulation format(MF) and optical signal-to-noise ratio(OSNR) based on machine learning have been gradually applied.However,existing methods have low accuracy for OSNR monitoring in complex scenarios because they do not consider the fine-grained characteristics of the signal.This paper proposes a joint monitoring model(FGNet) for optical signal MF and OSNR based on fine-grained constellation identification to solve this problem.Firstly,the backbone feature extraction module uses a deep residual structure.Secondly,a multilayer bilinear pooling module is proposed to perform fine-grained feature analysis on constellation features.Finally,a joint MF and OSNR monitoring module is proposed to realize the feature fusion of MF and OSNR.Extensive experiments with 7 200 constellation maps in the simulation dataset show that the proposed model has achieved superior performance compared to existing methods.

The resource scheduling strategy of container cloud system plays an important role in resource utilization and cluster performance.The existing container cluster scheduling does not fully take into account the resource occupancy within and between nodes,which is prone to container resource bottlenecks,resulting in low resource utilization and poor service reliability.In order to balance the workload of container cluster and reduce the bottleneck of container resources,this paper proposes a container cluster scheduling optimization algorithm CS-DQN(container scheduling optimization strategy based on DQN)based on deep Q-lear-ning network(DQN).Firstly,an optimization model of container cluster resource utilization for load balancing is proposed.Then,using the deep reinforcement learning method,a container cluster scheduling algorithm based on DQN is designed,and the relevant state space,action space and reward function are defined.By introducing the improved DQN algorithm,the container dynamic scheduling strategy which meets the optimization goal is generated based on the self-learning method.The prototype experimental results show that the scheduling strategy expands the scale of deployable containers in scheduling,achieves better load balancing in different workloads,improves resource utilization,and the service reliability is better guaranteed.

With the advent of the era of big data,the demand of large-scale computing makes the requirements on the performance of chips constantly increasing.Network-on-Chip(NoC),as a network-communication-centered interconnection structure on chips,has achieved a good tradeoff in all aspects of communication.The physical layout and interconnection mode of NoC components have a significant impact on the overall performance of the chip(such as signal delay,circuit cost).Due to the limited area of the chip,minimizing the total length of wires connecting components,in other words,minimizing wirelength,is considered as the key of chip design.The hierarchical cubic network is an excellent interconnection network which has less communication delay,better reliability and greater scalability while the linear array is one of the common topologies of NoC.When the hierarchical cubic network is ported to the linear array,the structure and algorithm of hierarchical cubic network can be simulated on the linear array.Graph embedding is a key technology to realize network porting.In graph embedding,the goal of minimizing the total wire length can be achieved by finding the optimal embedding with minimum wirelength.This paper mainly studies the optimal embedding problem of hierarchical cubic networks into linear arrays with minimum wirelength.Firstly,by studying the optimal set of the hierarchical cubic network,an embedding scheme hel of hierarchical cubic networks into linear arrays is proposed,and it is proved that the wirelength under hel is minimum compared with other embedding schemes,that is, hel is an optimal embedding.Then,the exact value of the wirelength under hel and an embedding algorithm with O(N) time complexity are given,where N=22n is the number of vertices of the n-dimensional hierarchical cube network.Furthermore,an algorithm of physical linear layout in NoC for hierarchical cubic networks is proposed.Finally,comparison experiments are conducted to evaluate the performance of embed-ding hel.

In the past decade,widespread data collection has become the norm.With the rapid development of large-scale data analysis and machine learning,data privacy is facing fundamental challenges.Exploring the trade-offs between privacy protection and data collection and analysis is a key scientific question.Differential privacy has become a de facto data privacy standard and has been widely studied and applied.Differential privacy technology can provide strict privacy protection for user data through a certain randomization mechanism.This paper provides a comprehensive overview of differential privacy technology and a summary and analysis of the latest progress of differential privacy.Specifically,this paper first gives a theoretical summary of differential privacy,including the central model,the local model and the shuffle model proposed in recent years.The three models are compared,and the advantages and disadvantages of different models are analyzed.Then,on the basis of the three models,some typical differential privacy mechanisms in literatures are analyzed from the perspective of algorithms.Then the current application of differential privacy technology in various fields is introduced.Finally,some new research topics about differential privacy technology are introduced,which expand the rich research direction for differential privacy technology.

IOS-XE network operating system is widely used in Cisco core routing and switching nodes,and its security is very important.However,its design focuses on the traffic fast-forwarding function and ignores protection for its own security which makes it faces great risks.In addition,the existing intrusion detection methods for traditional IOS system have problems such as poor real-time performance,inaccurate detection results and incomplete detection coverage when transplanted to the IOS-XE system.In order to strengthen the security of the IOS-XE system,this paper proposes a container-based intrusion detection method for Cisco IOS-XE system which can monitor the router states and requests in real time by deploying a detection container on the router.It solves the problems of configuration hidden attack detection,router https control traffic decryption and router state real-time monitor,which helps to detect the intrusion behavior of IOS-XE in real time.Experimental results show that this method can effectively detect common attacks against IOS-XE routers,including password guessing,Web injection,CLI injection,configuration hidden and backdoor implantation.Compared with existing detection methods,the proposed method has higher real-time performance and accuracy,and effectively improves the defense capability of IOS -XE routing devices.

As a typical representative of the Internet of Things,the Internet of Vehicles plays an important role in intelligent transportation,not only providing vehicles with a variety of online services,but also reducing the risk of accidents for drivers.However,a large number of sensitive information such as vehicle location and route are generated in the communication process of Internet of Vehicle networks.How to improve the anonymity of vehicle identity in the authentication service is a research hotspot in the field of Internet of Vehicles security.In this paper,an anonymous authentication scheme based on a batch authentication algorithm is proposed,which extends the anonymous authentication methods for IEEE WAVE security services through anonymous credentials,zero-knowledge proof and other technologies,and provides a method to recover identity through a trusted third party.Experimental results show that the computational cost of the scheme is better than that of the partial comparison scheme when the number of batch validated signatures exceeds 11.Besides,the optimal cycle of batch verification in BSM application of DSRC and vehicle near field payment application is also analyzed.

With the development of blockchain technology,smart contracts have been widely used in various fields,and Ethereum has become the largest smart contract platform.At the same time,the frequent smart contract vulnerabilities have caused huge economic losses.The vulnerability detection of smart contract has become the focus of research,while the previous smart contract vulnerability detection tools can not make good use of the syntax information of the contract source code.Aiming at the re-entrancy vulnerability of smart contract,firstly,this paper proposes SCDefender,a vulnerability detection tool based on deep learning.Taking the abstract syntax tree form of the Solidity source code of smart contract as the research object,the tree-based convolutional neural networks is used for vulnerability detection.Secondly,an abstract syntax tree pruning algorithm is proposed to remove the nodes irrelevant to the vulnerability detection task and retain the key information in the abstract syntax tree.The accuracy,recall rate and F1 value of SCDefender vulnerability detection is 81.43%,92.12% and 86.45% respectively,which has a good vulnerability detection effect.Ablation experiments show that the abstract syntax tree pruning algorithm has an important contribution to the vulnerability detection task of SCDefender.

Cisco’s new operating system,Cisco IOS-XE,is widely deployed on platforms such as Cisco routers and switches.However,there are vulnerabilities in the system’s Web management interface to allow permission escalation through command injection.Network security is facing serious threats.In recent years,fuzzing is usually used to detect security vulnerabilities in embedded devices,but there is currently no fuzzing framework for Cisco IOS-XE,and current fuzzing methods for IoT have poor performance due to the unique system architecture and command mode of IOS-XE.To solve the problems mentioned above,this paper proposes a novel fuzzing framework CRFuzzer for the Web management service in Cisco IOS-XE system to detect command injection vulnerabilities.CRFuzzer combines front-end requests and back-end scripts analysis to optimize seed generation,and locates vulnerable code based on characteristics of command injection to narrow the scope of testing.In order to evaluate the vulnerability detection performance of CRFuzzer,124 firmwares of 31 different versions are tested on the physical router ISR 4000 series and the cloud router CSR 1000v,and a total of 11 command injection vulnerabilities are detected,and 2 of them are undisclosed vulnerabilities.

APT(advanced persistent threat)are advanced persistent cyber-attack by hacker organizations to breach the target information system.Usually,the APTs are characterized by long duration and multiple attack techniques,making the traditional intrusion detection methods ineffective.Most existing APT detection systems are implemented based on manually designed rules by referring to domain knowledge(e.g.,ATT&CK).However,this way lacks of intelligence,generalization ability,and is difficult to detect unknown APT attacks.Aiming at this limitation,this paper proposes an intelligent APT detection method based on provenance data and graph neural networks.To capture the rich context information in the diversified attack techniques of APTs,it firstly models the system entities(e.g.,process,file,socket)in the provenance data into a provenance graph,and learns a semantic vector representation for each system entity by heterogeneous graph learning model.Then,to solve the problem of graph scale explosion caused by the long-term behaviors of APTs,APT detection is performed by sampling a local graph from the large scale heterogeneous graph,and classifying the key system entities as malicious or benign by graph convolution networks.A series of experiments are conducted on two datasets with real APT attacks.Experiment results show that the comprehensive performance of the proposed method outperforms other learning based detection models,as well as the state-of-the-art rule based APT detection systems.

Biomarker recognition is a major basis for achieving precision medicine,which plays an important role in diagnosing complex diseases,judging disease stages and evaluating the safety and effectiveness of new drugs or therapies in the target population.As the key technology of intelligent identification of biomarkers,the prediction of the association between circular RNA and disease is the key to deeply evaluate and measure the biological process,pathological process and intervention pathological response of subjects,and is one of the effective means and approaches to practice “precision medicine”.This paper comprehensively combs and prospects the circular RNA disease association prediction model in biological big data.Specifically,it first discusses the relationship between circRNA and disease in terms of the research background,physicochemical properties and functions of circ-RNAs.Then,it investigates the public database resources of circRNAs and diseases,and summarizes four computational methods of circRNA-disease association prediction from the perspective of computational models,and analyzes their advantages and shortcomings.Finally,this paper discusses the current challenges and future possible research directions of circRNA-disease association prediction problem.

Batched matrix computing problems are widely existed in scientific computing and engineering applications.With rapid performance improvements,GPU has become an important tool to solve such problems.The eigenvalue decomposition belongs to the two-sided decomposition and must be solved by the iterative algorithm.Iterative numbers for different matrices can be varied.Therefore,designing eigenvalue decomposition algorithms for batched matrices on the GPU is more challenging than designing batched algorithms for the one-sided decomposition,such as LU decomposition.This paper proposes batched algorithms based on the Jacobi algorithms for eigenvalue decomposition of Hermitian matrices.For matrices that cannot reside in shared memory wholly,the block technique is used to improve the arithmetic intensity,thus improving the use of GPU resources.Algorithms presented in this paper run completely on the GPU,avoiding the communication between the CPU and GPU.Kernel fusion is adopted to decrease the overhead of launching kernel and global memory access.Experimental results on V100 GPU show that our algorithms are better than existing works.Performance evaluation results of the Roofline model indicate that our implementations are close to the upper bound,approaching 4.11TFLOPS.

Aspect-based sentiment analysis(ABSA),as a downstream application of knowledge graph,belongs to the fine-grained sentiment analysis task,which aims to understand the sentiment polarity of people on the evaluation target at the aspect level.Relevant research in recent years has made significant progress,but existing methods focus on exploiting sequentiality or syntactic dependency constraints within sentences,and do not fully exploit the type of dependencies between context words and aspect words.In addition,the existing graph-based convolutional neural network models have insufficient ability to retain node features.In response to this problem,firstly,based on the syntactic dependency tree,this paper fully excavates the dependency types between context words and aspect words,and integrates them into the construction of the dependency graph.Second,we define a “sensitive relation set”,which is used to construct auxiliary sentences to enhance the correlation between specific context words and aspect words,and at the same time,combined with the sentiment knowledge network SenticNet to enhance the sentence dependency graph,and then improve the construction of the graph neural network.Finally,a context retention mechanism is introduced to reduce the information loss of node features in the multilayer graph convolution neural network.The proposed SS-GCN model fuses the syntactic and contextual representations learned in parallel to accomplish sentiment enhancement and syntactic enhancement,and extensive experiments on three public datasets demonstrate the effectiveness of SS-GCN.

Knowledge Graph-to-Text is a new task in the field of knowledge graph,which aims to transform knowledge graph into readable text describing these knowledge.With the deepening of research in recent years,the generation technology of Graph-to-Text has been applied to the fields of product review generation,recommendation explanation generation,paper abstract generation and so on.The translation model in the existing methods adopts the method of first-plan-then-realization,which fails to dynamically adjust the planning according to the generated text and does not track the static content planning,resulting in incohe-rent semantics before and after the text.In order to improve the semantic coherence of generated text,a Graph-to-Text model based on dynamic memory and two-layer reconstruction enhancement is proposed in this paper.Through three stages of static content planning,dynamic content planning and two-layer reconstruction mechanism,this model makes up for the structural difference between knowledge graph and text,focusing on the content of each triple while generating text.Compared with exis-ting generation models,this model not only compensates for the structural differences between knowledge graphs and texts,but also improves the ability to locate key entities,resulting in stronger factual consistency and semantics in the generated texts.In this paper,experiments are conducted on the WebNLG dataset.The results show that,compared with the current exis-ting models in the task of Graph-to-Text,the proposed model generates more accurate content planning.The logic between the sentences of the generated text is more reasonable and the correlation is stronger.The proposed model outperforms existing methods on me-trics such as BLEU,METEOR,ROUGE,CHRF++,etc.

The existing temporal knowledge graph completion models only consider the structural information of the quadruple itself,ignoring the implicit neighbor information and the constraints of relationships on entities,which leads to the poor perfor-mance of the models on the temporal knowledge graph completion task.In addition,some datasets exhibit unbalanced distribution in time,which makes it difficult for model training to achieve a good balance.To address these problems,the paper proposes a context-aware model based on relation constraints(CARC).CARC solves the problem of an unbalanced distribution of datasets in time through an adaptive time granularity aggregation module and uses a neighbor-aggregator to integrate contextual information into entity embeddings to enhance the embedding representation of the entity.In addition,the quadruple relation constraint mo-dule is designed to make the embeddings of entities with the same relational constraints close to each other,while those with diffe-rent relational constraints are far away from each other,which further enhances the embedding representation of entities.Extensive experiments are conducted on several publicly available temporal datasets,and the experimental results prove the superiority of the proposed model.

Named entity-relation joint extraction refers to extracting entity-relation triples from unstructured text.It's an important task for information extraction and knowledge graph construction.This paper proposes a new method--BERT-based global pointer network for named entity-relation joint extraction(BGPNRE).Firstly,the potential relation prediction module is used to predict the relations contained in the text,filters out the impossible relations,and limits the predicted relation subset for entity recognition.Then a relation-specific global pointer-net is used to obtain the location of all subject and object entities.Finally,a global pointer network correspondence component is designed to align the subject and object position into named entity-relation triples.This method avoids error propagation frompipeline model,and also solves the the redundancy of relation prediction,entity overlapping,and poor generalization of span-based extraction.Extensive experiments show that our model achieves state-of-the-art performance on NYT and WebNLG public benchmarks with higher performance gain on multi relations and entities overlapping.

Cross-item social recommendation is a method of integrating social relationships into the recommendation system.In social recommendation,user is the bridge connecting user-item interaction graph and user-user social graph.So user representation learning is essential to improve the performance of social recommendation.However,existing methods mainly use static attributes of users or items and explicit friend information in social networks for representations learning,and the temporal information of the interaction between users and items and their implicit friend information are not fully utilized.Therefore,in social recommendation,effective use of temporal information and social information has become one of the important research topics.This paper focuses on the temporal information of the interaction between users and items,and gives full play to the advantages of social network,modeling the user's implicit friends and item's social attributes.This paper proposes a novel graph neural networks social recommendation based on high-order and temporal features,referred to as HTGSR.Firstly,the framework uses gated recurrent unit to model item-based user representations to reflect the user's recent preferences,and defines a high-order mo-deling unit to extract the user's high-order connected features and obtain the user's implicit friend information.Secondly,HTGSR uses attention mechanism to obtain social-based user representation.Thirdly,the paper proposes different ways to construct item's social networks,and uses the attention mechanism to obtain item representations.Finally,the user's and item's representations are input to the MLP to complete the user's rating prediction for the item.The paper conducts specific experiments on two public and real-world datasets,and compares the experimental results with different recommendation algorithms.The results show that the HTGSR has achieved good results on the two datasets.

With the expansion of the scope in informatization fields,lots of text corpora in specific fields continue to appear.Due to the impact of security and sensitivity,the text corpora in these specific fields(e.g.,medical records corpora and communication corpora) are often small-scaled.It is difficult for traditional word embedding learning methods to obtain high-quality embeddings on these corpora.On the other hand,there may exist many out-of-vocabulary words in these corpora when using the existing pre-training language models directly,for which,many words cannot be represented as vectors and the performance on downstream tasks are limited.Many researchers start to study how to infer the semantics of out-of-vocabulary words and obtain effective out-of-vocabulary word embeddings based on fine-grained semantic information.However,the current models utilizing fine-grained semantic information mainly focus on the English corpora and they only model the relationship among fine-grained semantic information by simple ways of concatenation or mapping,which leads to a poor model robustness.Aiming at addressing the above problems,this paper first proposes to construct a fine-grained knowledge graph by exploiting Chinese word formation rules,such as the characters contained in Chinese words,as well as the character components and pinyin of Chinese characters.The know-ledge graph not only captures the relationship between Chinese characters and Chinese words,but also represents the multiple and complex relationships between Pinyin and Chinese characters,components and Chinese characters,and other fine-grained semantic information.Next,the relational graph convolution operation is performed on the knowledge graph to model the deeper relationship between fine-grained semantics and word semantics.The method further mines the relationship between fine-grained semantics by the sub-graph readout,so as to effectively infer the semantic information of Chinese out-of-vocabulary words.Experimental results show that our model achieves better performance on specific corpora with a large proportion of out-of-vocabulary words when applying to tasks such as word analogy,word similarity,text classification,and named entity recognition.

Knowledge graphs describe objective knowledge in the real world in a structured form,and are confronted with issues of completeness and newly-added knowledge.As an important means of complementing and updating knowledge graphs,know-ledge graph reasoning aims to infer new knowledge based on existing knowledge.In recent years,the research on knowledge graph reasoning based on repre-sentation learning has received extensive attention.The main idea is to convert the traditional reasoning process into semantic vector calculation based on the distributed representation of entities and relations.It has the advantages of fast calculation efficiency and high reasoning performance.In this paper,we review the knowledge graph reasoning based on repre sentation learning.Firstly,this paper summarizes the symbolic representation,data set,evaluation metric,training method,and evaluation task of knowledge graph reasoning.Secondly,it introduces the typical methods of knowledge graph reasoning,including translational distance and semantic matching methods.Thirdly,multi-source information fusion-based knowledge graph reasoning methods are classified.Then,neural network-based reasoning methods are introduced including convolutional neural network,graph neural network,recurrent neural network,and capsule network.Finally,this paper summarizes and forecasts the future research direction of knowledge graph reasoning.

Automatic electrocardiogram(ECG) abnormality detection is a multi-label classification problem,which is commonly solved by training a binary-relevance classifier for each abnormality.Due to the large number of abnormalities,the complex correlations between features and abnormalities,and those among different abnormalities,existing methods’ performance is not satis-fying.To make full use of the dependencies between features and abnormalities,this paper proposes a novel abnormality detection method based on label co-occurrence and feature local pertinence(LCFP).Firstly,we set up a consolidated feature space consisting of both the macro-features and micro-features based on the label co-occurrence and features’ pertineance.The macro-features are constructed with a clustering approach based on Dirichlet process mixture model(DPMM),thus distinguishing differentco-occurrence label sets.The micro-features are a subset of primitive features,which serves to distinguish between the labels in the same labelset.Next,we train a one-versus-all classifier which returns a relevance probability.Secondly,to make use of the diffe-rent correlation degrees among different abnormalities,we propose to differ the relevant labels from the irrelevant ones based on the sorting according to the probabilities given by the classifiers.In particular,we propose to exploit the Beta distribution to adaptively learn the anchor thresholds and correlation thresholds,thus determining the relevant labels of an instance.Our LCFP me-thod is a universal way to detect every possible ECG abnormalities,which effectively improves the detection accuracy.The results on two real datasets show that our method can achieves an improvement of 4% and 22.4%,respectively,in terms of F1,which proves the effectiveness of our method.

Sequential pattern mining with gap constraints is a special form of sequential pattern mining,which can reveal frequent subsequences in a certain gap.However,the current sequential pattern mining methods with gap constraints only focus on positive sequential pattern mining,and ignore the missing behavior in a series of events.To solve this problem,a negative sequential pattern method with periodic gap constraints(NSPG) mining is explored,which can reflect the relationship between elements more flexibly.To solve the problem of NSPG mining,this paper proposes an NSPG-INtree(incomplete nettrees) algorithm,which includes two key steps:candidate pattern generation and support calculation.For candidate pattern generation,to reduce the number of candidate patterns,the algorithm uses a pattern join strategy.For support calculation,to improve the efficiency and reduce space consumption,the algorithm employs an incomplete nettree structure to calculate the supports of patterns.Experimental results show that NSPG-INtree not only has high mining efficiency,but also can mine positive and negative sequential patterns with gap constraints.NSPG-INtree can find 209% ~ 352% more patterns than other gap-constrained sequential pattern mining algorithms.Moreover,NSPG-INtree can reduce the running time by 6%~38% than other competitive algorithms with different stra-tegies.

The ranking of nodes’ influence has always been a hot issue in the research area of complex networks.Susceptible-infected-recovered(SIR) model is an ideal nodes’ influence ranking method,which is commonly used to evaluate other nodes’ in-fluence ranking methods.But it is difficult to be applied in practice due to its high time complexity.This paper proposes a nodes’ influence ranking model based on sir value learning.Both the local structure and global structure information of nodes are used as features in the model.The sir value learning model is constructed by means of a deep learning model,which is trained on nodes’ features and sir data set in synthetic graphs with the same size.The trained model can predict sir value based on nodes’ features,and then rank nodes’ influence based on predicted sir.In this paper,a specific nodes’ influence ranking method is implemented based on the proposed model,and experiments are carried out on five real networks to verify the effectiveness of the method.The results show that the accuracy and monotonicity of nodes’ influence ranking results are improved compared with degree centrality,Kshell and Weighted Kshell degree neighborhood.

Much work has long been devoted to plant leaf classification,but these methods cannot be applied well in real applications,though they may achieve good results in public datasets.Moreover,they are hardly employed to more complex problems,e.g.leaf ranking,which requires the classification of leaves first and then ranking leaves of the same class.This paper proposes a new model for plant leaf classification as well as leaf ranking,which focuses on the granularity information of leaves and integrates multi-level granularity from coarse to fine.Specifically,the model contains two branches,coarse-grained and fine-grained,which are linked by a coarse-fine hybrid loss,prompting the model to progressively learn a coarse-to-fine representation.A multi-step training approach is used,with different levels of features extracted at each step,therefore enabling the fusion of shallow features with deep features.In addition,a geometric channel attention module,which consists of a spatial transformation and a bili-near attention pooling module,is proposed to allow our model to focus on more discriminative local regions in the image and extract more discriminative features.Our method achieves 99.8% and 99.7% classification accuracy on two publicly available leaf classification datasets,Flavia leaf and Swedish leaf,respectively,and 71.9% classification accuracy on our constructed tobacco leaf ranking dataset,both outperform the state-of-the-art methods.

The existing oil painting art style classification algorithms ignore the influence of the main area and the overall effect on the art style.Aiming at this problem,this paper proposes a new oil painting classification algorithm based on multi-feature fusion classifier(MFFC).Firstly,based on the common arrangement form of oil painting art elements,this paper designs the overlapping image block method.This method extracts spatial features of oil paintings to make up for the lack of composition style in existing algorithms.And it also can be used to distinguish the subject area from the background area.Secondly,the spatial features and the underlying features are combined in series to increase the location information of the elements in the picture.Finally,the spatial voting method is designed to give priority to the classification result of the main area as the output result of the algorithm.This is to highlight the role of oil painting subject area in the classification and realize the automatic classification of oil painting art style.Tested on the data set created by the FS-classifier model,its accuracy,precision,recall,F1-score and AUC reaches 96.92%,63.69%,98.75%,98.57% and 0.917,respectively.Compared with FS-classifier,the result increases by 6.72%,5.85%,9.05%,7.1% and 0.128,respectively.When tested on WIKIART and compared with other six algorithms,the accuracy improves by 13.27%,at least.The results show that the proposed algorithm can effectively improve the performance of spatial features for oil painting art style classification task,and has good practical value.

Evolutionary computation utilizes natural selection mechanisms and genetic laws in the process of biological evolution to solve optimization problems.The accuracy and efficiency of the evolutionary recurrent neural network model depends on the optimization effect of parameters and the structures.The utilization of evolutionary computation to solve the problem of adaptive optimization of parameters and structures in recurrent neural networks is a hot spot of automated deep learning.This paper summarizes the algorithms that combine evolutionary algorithms and recurrent neural networks.Firstly,it briefly reviews the traditional categories,common algorithms,and advantages of evolutionary computation.Next,it briefly introduces the structures and characteristics of the recurrent neural network models and analyzes the influencing factors of recurrent neural network perfor-mance.Then,it analyzes the algorithmic framework of evolutionary recurrent neural networks,and the current research development of evolutionary recurrent neural networks from weight optimization,hyperparameter optimization and structure optimization.Besides,other work on evolutionary recurrent neural networks is analyzed.Finally,it points out the challenges and the deve-lopment trend of evolutionary recurrent neural networks.

Recently,knowledge base(KB) has been widely adopted to the task of question answering(QA) to provide a proper answer for a given question,known as the KBQA problem.However,knowledge base itself may be incomplete(e.g.KB does not contain the answer to the question,or some of the entities and relationships in the question),limiting the overall performance of existing KBQA models.To resolve this issue,this paper proposes a new model to leverage textual documents for KBQA task by providing additional answers to enhance knowledge base coverage and background information to enhance the representation of questions.Specifically,the proposed model consists of three modules,namely entity and question representation module,document and enhanced-question representation module and answer prediction module.The first module aims to learn the representations of entities from the retrieved subgraph of knowledge base.Then,the question representation can be updated with the fusion of seed entities.The second module attempts to learn a proper representation of the document that is relevant to the given question.Then,the question representation can be further improved by fusing the document information.Finally,the last module makes an answer prediction based on the information of knowledge base,updated question and documents.Extensive experiments are conducted on the WebQuestionsSP dataset,and the results show that better accuracy can be obtained in comparison with other counterparts.

Compared with English text which is naturally composed of words,Chinese text has no word delimiters,so the combination of Chinese characters is more flexible,and it's more difficult to determine the entity boundaries in Chinese named entity recognition(NER).Current mainstream methods transform the NER task into a sequence labeling task.This paper studies the predicted label sequence under the BIOES tag scheme and calculates the entity boundary accuracy by separately considering the entity head label B or tail label E,which shows that increasing the boundary accuracy can further improve the accuracy of entity recognition.We expand the boundaries of entities with continuous labels,use the label type of the last character of the entity to correct the entity type,and use the word segmentation information to fill in the entity with incomplete labels.Finally,this paper proposes a BIO+ES labeling scheme that adds boundary labels to distinguish non-entity characters at entity boundaries and further improves the performance of Chinese NER.

Due to the diversity of Chinese characters and the complexity of Chinese semantic expressions,Chinese spelling che-cking is still an important and challenging task.Existing solutions usually suffer from the inability to dig deeper into the text semantics and often learn the mapping relationship between incorrect and correct characters through pre-established external resources or heuristic rules when exploiting the unique similarity features of Chinese characters.This paper proposes an end-to-end Chinese spelling checking algorithm model BFMBERT(BiGRU-Fusion Mask BERT) that incorporates multi-feature embedding of Chinese characters.The model first uses a pre-training task combining confusion sets to make BERT learn Chinese spelling error knowledge.It then employs a bi-directional GRU network to capture the probability of error for each character in the text.Furthermore,it applies this probability to compute a fusion embedding incorporating semantic,pinyin,and glyph features of Chinese characters.Finally,it feeds this fusion embedding into a mask language model in BERT to predict correct characters.BFMBERT is evaluated on the SIGHAN 2015 benchmark dataset and achieves an F1 value of 82.2,outperforming other baseline models.

Dependency parsing aims to analyze the syntactic structure of sentences from the perspective of linguistics.Existing studies suggest that combining such graph-like data with graph convolutional network(GCN) can help model better understand the text semantics.However,when dealing with dependency syntactic information as adjacency matrix,these methods ignore the types of syntactic dependency tags and the word semantics related to the tags,which makes the model unable to capture the deep emotional features.To solve the preceding problem,this paper proposes a Chinese short text sentiment analysis model CDSI(context and dependency syntactic information).This model can use BiLSTM(bidirectional long short-term memory) network to extract the context semantics of the text.Moreover,a dependency-aware embedding representation method is introduced to mine the contribution weights of different dependent paths to the sentiment classification task based on the syntactic structure.Then the GCN is used to model the context and dependent syntactic information at the same time,so as to strengthen the emotional features in the text representation.Based on SWB,NLPCC2014 and SMP2020-EWEC datasets,experimental results show that CDSI can effectively integrate the semantic and structural information in sentences,which achieves good results in both the Chinese short text sentiment binary classification and multi-classification tasks.

Automated essay scoring is a task that replaces manual grading for students’ essays,where rich semantics,rigorous organization,and reasonable logic are important considering factors.Most previous studies only consider the semantics or organization of the essay from a single perspective,lacking considering higher-level factors such as logic.Therefore,this paper proposes a multi-perspective evaluation framework(MPE) to more objective and reliable evaluate the essay from semantics,organization,and logic.MPE first utilizes the pre-trained model to encode sentence and obtain three levels semantic information to evaluate the essay's semantic expression.Then,it combines sentence function identification and paragraph function identification to evaluate the essay′s organization.Moreover,MPE evaluates the essay's logic by calculating the coherence between paragraphs.Finally,the framework scores the essay by integrating these three evaluation perspectives.Experimental results show that the proposed multi-perspective evaluation framework can effectively score the essays at various qualities,outperforming all the baselines.

Real-time collision avoidance in dynamic environments is a challenge in trajectory planning of mobile robots. Focusing on environments with variable number of obstacles,this paper proposes a real-time trajectory planning algorithm,Crit-LSTM-DRL,based on long short-term memory(LSTM) and deep reinforcement learning(DRL). First,it predicts the time to the occurrence of a collision between an obstacle and the robot based on their states,and then computes the collision criticality of each obstacle with respect to the robot. Second,it generates the obstacle sequence based on the collision criticality and abstracts a fixed-dimension vector by LSTM to represent the environment. Finally,the robot state and the extracted vector are concatenated as the input of the DRL's value network to compute the value with respect to the system state. At any instant,for each action,it predicts the value of the next state based on the LSTM and DRL models and then the value of the current state; hence,the action generating the maximal value of the current state is selected to control the robot. To evaluate the performance of Crit-LSTM-DRL,it is first trained in three different environments and obtain three models: the model trained in the environment with 5 obstacles,the model trained in the environment with 10 obstacles,and the model trained in the environment with variable number of obstacles(1～10). The models then are tested in various environments containing different number of obstacles. To further investigate the effects of the interaction between an obstacle and the robot,this paper also takes the joint state of an obstacle and the robot as the state of the obstacle and trains another three models in the above training environments. Experimental results show the effectiveness and efficiency of Crit-LSTM-DRL.

In recent years,artificial intelligence represented by deep learning has made breakthroughs in theories and technologies.With the strong support of data,algorithms and computing power,deep learning has received unprecedented attention and has been widely used in various fields,bringing great improvements to the corresponding fields.With the wide application of deep learning technology in various fields including security critical ones,the security issue of deep learning has attracted more and more attention.Researchers have found many security risks in deep learning systems.In terms of the security of deep learning models,researchers have extensively explored the new attack paradigm of backdoor attack.Backdoor attack can threaten deep learning models throughout their whole life cycle.A large number of researchers have proposed series of attack scheme from different angles.This paper takes the security threats of deep learning system as a starting point,introduces the current attack paradigms.On this basis,it gives the back-ground and principle of backdoor attack,distinguishes the similar attack paradigms such as adversarial attack and data poisoning attack,then continues to elaborate on the attack principle and outstanding features of the classic methods of backdoor attack to date.According to the working principle,the attack schemes are divided into data poisoning based attack and model poisoning based attack and others,the paper systematically summarizes them and clarify the advantages and disadvantages of current research.Then,this paper surveys the state-of-the-art works of backdoor attack against various typical applications and popular deep learning paradigms,which further reveal the threat of backdoor attack towards deep learning models.Finally,this paper summarizes the research work on applying backdoor attack characteristics to positive applications and explores the current challenges of backdoor attack,as well as discusses future research directions worthy of in-depth exploration,aiming to provide guidance for the follow-up researchers to further promote the development of backdoor attack and security of deep learning.

In recent years,machine learning has not only achieved remarkable results in conventional fields such as computer vision and natural language processing,but also been widely applied to process sensitive data such as face images,financial data and medical information.Recently,researchers find that machine learning models will remember the data in their training sets,making them vulnerable to membership inference attacks,that is,the attacker can infer whether the given data exists in the training set of a specific machine learning model.The success of membership inference attacks may lead to serious individual privacy leakage.For example,the existence of a patient's medical record in a hospital's analytical training set reveals that the patient was once a patient there.The paper first introduces the basic principle of membership inference attacks,and then systematically summarizes and classifies the representative research achievements on membership inference attacks and defenses in recent years.In particular,how to attack and defend under different conditions is described in detail.Finally,by reviewing the development of membership inference attacks,this paper explores the main challenges and potential development directions of machine learning privacy protection in the future.

Traditional IP-based Internet offers an end-to-end data transport service and has developed rapidly in the past half-century.However,serious security incidents emerged from attacks based on traditional networks.Traditional security mechanisms(e.g.,firewalls,intrusion detection systems) enhance security.However,most of them only provide some remedial strategies rather than solve the address-security problem radically due to the lack of change in network design.The overall in-depth security of the networked system cannot be guaranteed without a fundamental change.In order to meet the development requirements of the next generation of an endogenous security network,one of the future networks,the multi-identifier network(MIN),is introduced as our research background.This paper proposes an efficient scheme in hieratical architecture that provides comprehensive protection by addressing the security aspects pertaining to the network and application layers.At the network layer,the proposed architecture develops a multi-identifier routing scheme with embedded identity-based authentication and packet signature mechanisms to provide data tamper-resistance and traceability.At the application layer,the proposed architecture designs a mimic defensive scheme combined with weighted network centrality measures.This scheme focuses on protecting the core components of the whole network to improve the service's robustness and efficiently resist potential attacks.This paper tests and evaluates the proposed scheme from a theoretical and practical perspective.An analytical model is built based on the random walk for theoretical evaluation.In experiments,the proposed scheme is developed in MIN as MIN-VPN.Then considering IP-VPN as a baseline,anti-attack tests are conducted on IP-VPN and MIN-VPN.The results of theoretical evaluations and experiments show that the proposed scheme provides excellent transmission performance and successful defense against various TCP/IP-based attacks with acceptable defensive cost,demonstrating this security mechanism's effectiveness.In addition,after long-period penetration testing in three international elite security contests,the proposed method is effectively immune to all TCP/IP-based attacks from thousands of professional teams,thus verifying its strong security.

With the rapid development of the Internet,the security of network equipment has received extensive attention.Aiming at the problems of that the existing network equipment anomaly detection technology is destructive and difficult to detect,the paper uses the packets delay spent by the network equipment to transmit and process data packets as the detection basis,and proposes an anomaly detection scheme based on delay characteristics.The proposed scheme adopts side channel analysis,and it does not need to upgrade the equipment's software or hardware.It has the characteristics of non-intrusive and easy to implement.Firstly,the method uses the high-precision timing technology time stamp machine to collect the time delay information,and uses the genetic algorithm to extract the peak position feature of the delay distribution.Secondly,to solve the problem of the imbalance of data set,the method uses one-class support vector machine algorithm to construct anomaly detection algorithm.Finally,the validity of the method is verified by building an experimental platform,and the experimental results are evaluated.Experimental results show that the proposed method is feasible and effective.

Intrusion detection system(IDS) is a detection system that can issue an alarm when a network attack occurs.Detecting unknown attacks in the network is a challenge that IDS faces.Deep learning technology plays an important role in network traffic anomaly detection,but most of the existing methods have a high false positive rate and most of the models are trained using supervised learning methods.A gated recurrent unit network(GRU)-based semi-supervised network traffic anomaly detection me-thod(SEMI-GRU) is proposed,which combines a multi-layer bidirectional gated recurrent unit neural network(MLB-GRU) and an improved feedforward neural network(FNN).Data oversampling technology and semi-supervised learning training method are used to test the effect of network traffic anomaly detection using binary classification and multi-classification methods,and NSL-KDD,UNSW-NB15 and CIC-Bell-DNS-EXF-2021 datasets are used for verification.Compared with classic machine learning mo-dels and deep learning models such as DNN and ANN,the SEMI-GRU method outperforms the machines lear-ning and deep learning methods listed in this paper in terms of accuracy,precision,recall,false positives,and F1 scores.In the NSL-KDD binary and multi-class tasks,SEMI-GRU outperforms other methods on the F1 score metric,which is 93.08% and 82.15%,respectively.In the UNSW-NB15 binary and multi-class tasks,SEMI-GRU outperforms the other methods on the F1 score,which is 88.13% and 75.24%,respectively.In the CIC-Bell-DNS-EXF-2021 light file attack dataset binary classification task,all test data are classified correctly.