Started in January,1974(Monthly)
Supervised and Sponsored by Chongqing Southwest Information Co., Ltd.
ISSN 1002-137X
CN 50-1075/TP
CODEN JKIEBK
Editors
Current Issue
Volume 50 Issue 4, 15 April 2023
  
Database & Big Data & Data Science
Review of Deep Learning Applications in Healthcare
XUE Fenghao, JIANG Haibo, TANG Dan
Computer Science. 2023, 50 (4): 1-15.  doi:10.11896/jsjkx.220600166
Abstract PDF(2832KB) ( 623 )   
References | Related Articles | Metrics
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.
Study on Degree of Node Based Personalized Propagation of Neural Predictions forSocial Networks
SHAO Yunfei, SONG You, WANG Baohui
Computer Science. 2023, 50 (4): 16-21.  doi:10.11896/jsjkx.220300274
Abstract PDF(1783KB) ( 521 )   
References | Related Articles | Metrics
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.
Dual-attention Network Model on Propagation Tree Structures for Rumor Detection
HAN Xueming, JIA Caiyan, LI Xuanya, ZHANG Pengfei
Computer Science. 2023, 50 (4): 22-31.  doi:10.11896/jsjkx.220200037
Abstract PDF(2510KB) ( 420 )   
References | Related Articles | Metrics
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.
LayerLSB:Nearest Neighbors Search Based on Layered Locality Sensitive B-tree
DING Jiwen, LIU Zhuojin, WANG Jiaxing, ZHANG Yanfeng, YU Ge
Computer Science. 2023, 50 (4): 32-39.  doi:10.11896/jsjkx.220600078
Abstract PDF(3236KB) ( 411 )   
References | Related Articles | Metrics
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.
Short-time Traffic Flow Forecasting Based on Multi-adjacent Graph and Multi-head Attention Mechanism
YIN Heng, ZHANG Fan, LI Tianrui
Computer Science. 2023, 50 (4): 40-46.  doi:10.11896/jsjkx.220200079
Abstract PDF(2022KB) ( 376 )   
References | Related Articles | Metrics
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.
Sequential Recommendation Model Based on User’s Long and Short Term Preference
LUO Xiaohui, WU Yun, WANG Chenxing, YU Wenting
Computer Science. 2023, 50 (4): 47-55.  doi:10.11896/jsjkx.220100264
Abstract PDF(2846KB) ( 371 )   
References | Related Articles | Metrics
Aiming at the problem that the existing sequence recommendation model ignores the personalized behavior of different users,the model cannot fully capture the interest drift caused by users’ dynamic preferences,a sequence recommendation model based on users’ long and short term preferences(ULSP-SRM)is proposed.Firstly,the dynamic category embedding of the user is generated according to the category and time information of the interactive items in the user’s sequence,thereby effectively establishing the correlation between the items and reducing the sparsity of the data.Secondly,according to the time interval information of the user’s current clicked item and the last clicked item,a personalized time series position embedding matrix is generated to simulate the user’s personalized aggregation phenomenon and better reflect the dynamic change of user preference.Then,the user’s long-term preference sequence fused with the personalized time-series position embedding matrix is input into the gated recurrent unit in units of sessions to generate the user’s long-term preference representation,and the user’s long and short term preferences are fused through the attention mechanism to generate the final preference representation of the user,to achieve the purpose of fully capturing the user’s preference.Finally,the final user preference representation is input to the recommendation prediction layer for the next recommendation prediction.Experiments are carried out on seven subsets of Amazon public data set,and the area under curve(AUC ),recall rate and precision rate indicators are used for comprehensive evaluation.Experimental results show that the proposed model outperforms other advanced benchmark models,effectively improving recommended perfor-mance.
Community Detection Based on Markov Similarity Enhancement and Network Embedding
ZENG Xiangyu, LONG Haixia, YANG Xuhua
Computer Science. 2023, 50 (4): 56-62.  doi:10.11896/jsjkx.220100155
Abstract PDF(1995KB) ( 279 )   
References | Related Articles | Metrics
Community structure is ubiquitous in various complex networks in nature and is one of the important characteristics of network structure.Community detection can identify useful information in the network,and help to analyze the structure and function of the network.It is widely used in social networks,biology,medicine and other fields.Aiming at the low accuracy of the current community detection algorithm based on local similarity in complex networks,a community detection algorithm based on Markov similarity enhancement and network embedding is proposed.Firstly,inspired by the idea of Markov chain,a Markov similarity enhancement method is proposed,which obtains the steady-state Markov similarity enhancement matrix through the Mar-kov iterative state transition of the initial network.According to the Markov similarity index,the network is divided into initial community structure.Then,a new community similarity index is proposed by combining the network topology and network embedding.The small community in the initial community structure is merged with its closely connected community to obtain the network community structure.On 7 real networks and artificial networks with variable parameters,compared with other 5 well-known community detection algorithms,it is proved that the proposed algorithm has a good community detection effect.
Same Effect Relation and Concept Reduction in Formal Concept Analysis
MA Wensheng, HOU Xilin
Computer Science. 2023, 50 (4): 63-76.  doi:10.11896/jsjkx.221000169
Abstract PDF(3037KB) ( 263 )   
References | Related Articles | Metrics
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.
Computer Graphics & Multimedia
Deep Learning-based Visual Multiple Object Tracking:A Review
WU Han, NIE Jiahao, ZHANG Zhaowei, HE Zhiwei, GAO Mingyu
Computer Science. 2023, 50 (4): 77-87.  doi:10.11896/jsjkx.220300173
Abstract PDF(2496KB) ( 583 )   
References | Related Articles | Metrics
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 Examples Generation Method Based on Image Color Random Transformation
BAI Zhixu, WANG Hengjun, GUO Kexiang
Computer Science. 2023, 50 (4): 88-95.  doi:10.11896/jsjkx.211100164
Abstract PDF(5462KB) ( 286 )   
References | Related Articles | Metrics
Although deep neural networks(DNNs) have good performance in most classification tasks,they are vulnerable to adversarial examples,making the security of DNNs questionable.Research designs to generate strongly aggressive adversarial examples can help improve the security and robustness of DNNs.Among the methods for generating adversarial examples,black-box attacks are more practical than white-box attacks,which need to rely on model structural parameters.Black-box attacks are gene-rally based on iterative methods to generate adversarial examples,which are less migratory,leading to a generally low success rate of their black-box attacks.To address this problem,introducing data enhancement techniques in the process of countermeasure example generation to randomly change the color of the original image within a limited range can effectively improve the migration of countermeasure examples,thus increasing the success rate of countermeasure example black box attacks.This method is validated through adversarial attack experiments on ImageNet dataset with normal network and adversarial training network,and the experimental results indicate that the method can effectively improve the mobility of the generated adversarial examples.
Skin Lesion Segmentation Combining Boundary Enhancement and Multi-scale Attention
BAI Xuefei, JIN Zhichao, WANG Wenjian, MA Yanan
Computer Science. 2023, 50 (4): 96-102.  doi:10.11896/jsjkx.220300054
Abstract PDF(2952KB) ( 301 )   
References | Related Articles | Metrics
In view of the various types of skin lesions in shape,color and size,which pose a huge challenge to the accurate segmentation of skin lesions,a skin lesion segmentation network that combines boundary enhancement and multi-scale attention is proposed(BEMA U-Net).It consists of two modules,one is called spatial multi-scale attention module,which is used to extract spatial global features,and the other is called boundary enhancement module,which is used to enhance the edge features of the lesion area.BEMA U-Net adds the two modules to the U-Net network with encoding and decoding structure,which can effectively suppress the interference of background noise in the image of lesions and enhance the edge details of lesions.In addition,the mixed loss function is designed,Dice loss and Boundary loss are combined,and the dynamic weight adjustment of the mixed loss function is realized in the training process,so that the network could carry out multiple supervision on the extraction of the overall features and edge details of the pathological images,and the problems of hair interference and edge blur in the segmentation of skin pathological images are alleviated.Experimental results on ISIC2017 and ISIC2018 public data sets show that the proposed algorithm has better segmentation effect for skin lesions with continuous edges and clear contours.
Human Parsing Model Combined with Regional Sampling and Inter-class Loss
LI Yang, HAN Ping
Computer Science. 2023, 50 (4): 103-109.  doi:10.11896/jsjkx.220100259
Abstract PDF(3480KB) ( 233 )   
References | Related Articles | Metrics
Human parsing is a fine-grained level semantic segmentation task.The refinement of annotated categories in the human parsing dataset makes the dataset follow a long-tailed distribution and improves the difficulty of identifying similar categories.Balanced sampling is an efficient way to solve long-tailed distribution problem,but it’s difficult to achieve balanced sampling of the labeled object in human parsing.On the other hand,the fine-grained annotation will make the model misjudge similar categories.In response to these problems,a human parsing model combined with regional sampling and inter-class loss is proposed.The model consists of the semantic segmentation network,regionally balanced sampling module(RBSM),and inter-class loss module(ILM).Firstly,the images are parsed by the semantic segmentation network.Next,the parsing results and the ground truth labels are sampled by regionally balanced sampling module.Then the sampled parsing results and sampled ground truth labels are utilized to calculate the master loss.Meanwhile,the inter-class loss between the heatmap features coming from the semantic segmentation network and ground truth labels are calculated in the inter-class loss module,and the master loss and the inter-class loss are optimized at the same time to get a more accurate model.Experimental results based on the MHPv2.0 dataset show that the mIoU of the proposed model improves by more than 1.3% without changing the structure of the semantic segmentation network.The algorithm effectively reduces the impact of the long tail distribution problem and similarity among categories.
Adaptive Image Adversarial Reprogramming Based on Noise Invisibility Factors
LIU Yifan, OU Bo, XIONG Jianqin
Computer Science. 2023, 50 (4): 110-116.  doi:10.11896/jsjkx.220300024
Abstract PDF(3719KB) ( 280 )   
References | Related Articles | Metrics
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.
Image Compressed Sensing Attention Neural Network Based on Residual Feature Aggregation
WANG Zhenbiao, QIN Yali, WANG Rongfang, ZHENG Huan
Computer Science. 2023, 50 (4): 117-124.  doi:10.11896/jsjkx.211200215
Abstract PDF(3733KB) ( 267 )   
References | Related Articles | Metrics
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.
Numerical Solution of Saint-Venant Equation by Cubic B-spline Quasi-interpolation
QIAN Jiang, ZHANG Ding
Computer Science. 2023, 50 (4): 125-132.  doi:10.11896/jsjkx.220800118
Abstract PDF(3686KB) ( 275 )   
References | Related Articles | Metrics
Firstly,the error estimates of cubic spline quasi-intepolating operators are derived for continuous differential function with different orders.Secondly,cubic B-spline quasi-interpolation is used to get the numerical solution of Saint-Venant equation.Specifically,the derivatives of the quasi-interpolation are used to approximate the spatial derivative of the dependent variable and forward difference method is used to approximate the time derivative of the dependent variable.Finally,the numerical solutions are compared with the solution obtained by the fourth order Runge-Kutta method and the leapfrog scheme.Then numerical examples show that cubic spline quasi-intepolating method has some advantages.
Image Denoising Algorithm Based on Deep Multi-scale Convolution Sparse Coding
YIN Haitao, WANG Tianyou
Computer Science. 2023, 50 (4): 133-140.  doi:10.11896/jsjkx.220100090
Abstract PDF(13468KB) ( 325 )   
References | Related Articles | Metrics
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.
Text-Image Cross-modal Retrieval Based on Transformer
YANG Xiaoyu, LI Chao, CHEN Shunyao, LI Haoliang, YIN Guangqiang
Computer Science. 2023, 50 (4): 141-148.  doi:10.11896/jsjkx.220100083
Abstract PDF(2709KB) ( 329 )   
References | Related Articles | Metrics
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.
Artificial Intelligence
Comprehensive Survey of Loss Functions in Knowledge Graph Embedding Models
SHEN Qiuhui, ZHANG Hongjun, XU Youwei, WANG Hang, CHENG Kai
Computer Science. 2023, 50 (4): 149-158.  doi:10.11896/jsjkx.211200175
Abstract PDF(2702KB) ( 498 )   
References | Related Articles | Metrics
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.
Review of Intelligent Traffic Signal Control Strategies Driven by Deep Reinforcement Learning
YU Ze, NING Nianwen, ZHENG Yanliu, LYU Yining, LIU Fuqiang, ZHOU Yi
Computer Science. 2023, 50 (4): 159-171.  doi:10.11896/jsjkx.220500261
Abstract PDF(2715KB) ( 344 )   
References | Related Articles | Metrics
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.
Mixed-curve for Link Completion of Multi-relational Heterogeneous Knowledge Graphs
LI Shujing, HUANG Zengfeng
Computer Science. 2023, 50 (4): 172-180.  doi:10.11896/jsjkx.220500135
Abstract PDF(2479KB) ( 262 )   
References | Related Articles | Metrics
Knowledge graphs(KGs)has gradually become valuable asset in the field of AI.However,a major problem is that there are many missing edges in the existing KGs.KGs representation learning can effectively solve this problem.The quality of representation learning depends on how well the geometry of the embedding space matches the structure of the data.Euclidean space has been the main force for embeddings;hyperbolic andspherical spaces gaining popularity due to their ability to better embed new types of structured data.However,most data are highly heterogeneous,the single-space modeling leads to large information distortion.To solve this problem,inspired by MuRP model,mixed-curve space model is proposed to provide representations suitable for heterogeneous structural data.Firstly,the Descartes product of Euclidean hyperbolic and spherical spaces is used to construct mixed space.Then,a graph attention mechanism is designed to obtain the importance of relationship.Experimental results on three KGs benchmark datasets show that the proposed model can effectively alleviate the problems caused by heterostructural embedding in low-dimensional spaces with constant curvature.The proposed method is applied to the cold start problem of recommender system,and the corresponding indicators have been improved to a certain extent.
Incorporating Multi-granularity Extractive Features for Keyphrase Generation
ZHEN Tiange, SONG Mingyang, JING Liping
Computer Science. 2023, 50 (4): 181-187.  doi:10.11896/jsjkx.220700164
Abstract PDF(3147KB) ( 280 )   
References | Related Articles | Metrics
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.
Study on Extractive Summarization with Global Information
ZHANG Xiang, MAO Xingjing, ZHAO Rongmei, JU Shenggen
Computer Science. 2023, 50 (4): 188-195.  doi:10.11896/jsjkx.220200061
Abstract PDF(2237KB) ( 292 )   
References | Related Articles | Metrics
Extractive automatic text summarization aims to extract the sentences that can best express the semantics of the full text from the original text to form a summary.It is widely used and studied due to its simplicity and efficiency.Currently,extractive summarization models are mostly based on the local relationship between sentences to obtain importance scores to select sentences.This method ignores the global semantic information of the original text,and the model is more susceptible to the influence of local non-important relationships.Therefore,an extractive summarization model incorporating global semantic information is proposed.After obtaining the representation of sentences and articles,the model learns the relationship between sentences and global information through the sentence-level encoder and global information extraction module and then integrates the extracted global information into the sentence vector.Finally,the sentence score is obtained to determine whether it is a summary sentence.The proposed model can achieve end-to-end training,and two global information extraction techniques based on aspect extraction and neural topic model are studied in the global information extraction module.Experimental results on the public dataset CNN/DailyMail verify the effectiveness of the model integrating global information.
Aspect-level Sentiment Classification Based on Interactive Attention and Graph Convolutional Network
WANG Yali, ZHANG Fan, YU Zeng, LI Tianrui
Computer Science. 2023, 50 (4): 196-203.  doi:10.11896/jsjkx.220100105
Abstract PDF(2060KB) ( 369 )   
References | Related Articles | Metrics
Aspect-level sentiment analysis is a key task in fine-grained sentiment analysis,which aims to predict the sentiment tendency of different aspect terms in a sentence.In view of the fact that the current research combined with graph convolution network ignores the meaning of aspect terms themselves and the interaction between aspect terms and context,an interactive attention graph convolutional network model is proposed,named interactive attention graph convolution network(IAGCN).It firstlycombines BiLSTM and modified dynamic weights to model context.Secondly,the syntactic information is encoded by exploiting graph convolutional network on syntactic dependency tree.Then,the attention among context and aspect terms is investigated through interactive attention mechanism and the representation of context and aspect term is reconstructed.Finally,the sentiment polarity of a given aspect term is obtained through a softmax layer.Compared with the baseline models,the accuracy rate and F1 score of the proposed model improves by 0.56%~1.75% and 1.34%~4.04% on 5 datasets,respectively.At the same time,the pre-training model BERT is applied to this task.Compared with the IAGCN based on GloVe model,its accuracy rate and F1 score increases by 1.47%~3.95% and 2.59%~7.55%,respectively.Thus,the model effect has been further improved.
Chaotic Adaptive Quantum Firefly Algorithm
LIU Xiaonan, AN Jiale, HE Ming, SONG Huichao
Computer Science. 2023, 50 (4): 204-211.  doi:10.11896/jsjkx.220100242
Abstract PDF(1411KB) ( 254 )   
References | Related Articles | Metrics
In order to improve the search performance of quantum firefly algorithm(QFA) and solve the problem that it is easy to fall into local optimality when facing some problems,an improved QFA with chaotic map,neighborhood search and adaptive random disturbance is proposed,named chaos adaptive quantum firefly algorithm(CAQFA).In this algorithm,chaotic map is applied to the initialization stage of the population to improve the quality of the initial population.In the update stage,the neighborhood search is carried out for the optimal individual of the current population to enhance the ability of the algorithm to jump out of the local optimization.The introduction of adaptive random disturbance to other individuals increases the randomness of the algorithm and achieves a balance between the exploration and development of search space,so as to improve the performance of the algorithm.Eighteen different types of benchmark functions are selected to test the performance of the algorithm.The test results show that CAQFA has better search ability,stability and strong competitiveness compared with firefly algorithm(FA),QFA and quantum particle swarm optimization(QPSO).
Computer Network
Failure Recovery Model for Single Link with Congestion-Avoidance in SDN
CHEN Ziqiang, XIA Zhengyou
Computer Science. 2023, 50 (4): 212-219.  doi:10.11896/jsjkx.220300184
Abstract PDF(3300KB) ( 264 )   
References | Related Articles | 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.
Optical Performance Monitoring Method Based on Fine-grained Constellation Diagram Recognition
CHEN Jinjie, HE Chao, XIAO Xiao, LEI Yinjie
Computer Science. 2023, 50 (4): 220-225.  doi:10.11896/jsjkx.220600238
Abstract PDF(2975KB) ( 286 )   
References | Related Articles | Metrics
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.
Automatic Modulation Recognition Method Based on Multimodal Time-Frequency Feature Fusion
HE Chao, CHEN Jinjie, JIN Zhao, LEI Yinjie
Computer Science. 2023, 50 (4): 226-232.  doi:10.11896/jsjkx.220600242
Abstract PDF(3124KB) ( 310 )   
References | Related Articles | Metrics
Automatic modulation recognition (AMR) is a key technology in cognitive radio and has a wide range of applications in wireless communication.Aiming at the problem that most of the existing automatic modulation classification methods only use the single-modal information in the time domain or frequency domain,ignoring the complementarity between the multi-modal information,a signal modulation classification recognition method based on multimodal time-frequency feature fusion is proposed.First,the time-domain features and frequency-domain features of the signal are aligned by contrastive learning before fusion to reduce the heterogeneity difference.Secondly,cross-modal attention is used to achieve complementary fusion of time-domain features and frequency-domain features.Finally,in order to further improve the overall performance of the model,a residual shrin-kage module is introduced into the frequency domain encoder to extract the frequency domain features of the time-frequency map and the complex bidirectional gated recurrent unit is introduced into the time domain encoder to extract the correlation features between the I and Q signals and the time-domain features.Experimental results on RadioML2016a show that the proposed me-thod has higher recognition accuracy and noise robustness.
Self-balanced Scheduling Strategy for Container Cluster Based on Improved DQN Algorithm
XIE Yongsheng, HUANG Xiangheng, CHEN Ningjiang
Computer Science. 2023, 50 (4): 233-240.  doi:10.11896/jsjkx.220300215
Abstract PDF(3527KB) ( 279 )   
References | Related Articles | Metrics
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.
Auction-based Edge Cloud Deadline-aware Task Offloading Strategy
PEI Cui, FAN Guisheng, YU Huiqun, YUE Yiming
Computer Science. 2023, 50 (4): 241-248.  doi:10.11896/jsjkx.211200194
Abstract PDF(2613KB) ( 235 )   
References | Related Articles | Metrics
With the advent of the Internet of everything and the 5G era,the amount of data that mobile users need to process does not match their data processing capabilities,offloading a large number of tasks to limited edge servers for execution is bound to produce competition.The introduction of auction modelcan solve the problem of resource competition among users.At present,most task offloading works based on auction ignore the deadline perception of tasks,the general task offloading work only consi-ders delay-sensitive tasks,and does not consider ensuring the security of the offloading process.Therefore,an auction based deadline-aware task offloading(ABDTO) strategy is proposed,which uses the auction mechanism based on smart contract to realize the optimal allocation of deadline-aware tasks(delay-sensitive tasks and non-delay-sensitive tasks) to the edge ser-vers,and the total utility(i.e.total profit) is taken as the evaluationcriterion to achieve a win-win situation between mobile users and edge ser-vers.The heuristic genetic algorithm is used to conduct simulation experiments.Compared with TACD,UPPER and RND algorithms,ABDTO strategy has higher overall utility.Finally,using Remix,Ganache,etc.to establish the Ethereum private blockchain network for simulation,which proves the correctness and feasibility of the strategy.
Optimal Embedding of Hierarchical Cubic Networks into Linear Arrays of NoC
GUO Ruyan, WANG Yan, FAN Jianxi, FAN Weibei
Computer Science. 2023, 50 (4): 249-256.  doi:10.11896/jsjkx.220100019
Abstract PDF(2013KB) ( 246 )   
References | Related Articles | Metrics
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.
Deadline Constrained Scheduling Optimization Algorithm for Workflow in Clouds Using Spot Instance
PAN Jikui, DONG Xinyi, LU Zhenghao, WANG Zijian, SUN Fuquan
Computer Science. 2023, 50 (4): 257-264.  doi:10.11896/jsjkx.220100100
Abstract PDF(3310KB) ( 247 )   
References | Related Articles | Metrics
In recent years,due the advantages of on-demand resource provisioning and pay-as-you-go billing model,it is increa-singly popular to execute large-scale workflow applications in cloud environments.Cloud service providers offer resources with different capabilities at different prices.In order to improve resource utilization,many cloud service providers provide transient resources at a much lower price than normal resources.Spot instance provided by Amazon EC2 can greatly reduce the execution cost of workflow.One of the main problems of workflow scheduling in cloud is to find a cheaper scheduling method on the premise of meeting the deadline.To solve this problem,a deadline constrained scheduling optimization algorithm for workflow in clouds using spot instance(Spot-ProLis) is proposed.The algorithm takes into account the case that the data transmission time of the same virtual machine is zero,and uses the method of probabilistic upward rank to order tasks.In the resource allocation stage,spot instances are added as candidate resources,which effectively reduces the execution cost.Experiment results show that compared with the classical ProLis algorithm,Spot-ProLis has significant advantages in reducing the execution cost.
Information Security
Review of Differential Privacy Research
ZHAO Yuqi, YANG Min
Computer Science. 2023, 50 (4): 265-276.  doi:10.11896/jsjkx.220500292
Abstract PDF(1623KB) ( 429 )   
References | Related Articles | Metrics
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.
Research on PoC Refactoring of Third-party Library in Heterogeneous Environment
SONG Wenkai, YOU Wei, LIANG Bin, HUANG Jianjun, SHI Wenchang
Computer Science. 2023, 50 (4): 277-287.  doi:10.11896/jsjkx.220500092
Abstract PDF(6814KB) ( 271 )   
References | Related Articles | Metrics
Vulnerabilities in third-party libraries are widely propagated to host applications(software that using third-party libra-ries),and developers of host applications usually fail to fix these vulnerabilities in a timely manner,which easily leads to security problems.In order to explore the impact of third-party library vulnerabilities on the host applications,it is particularly important to effectively verify whether the vulnerabilities propagated to the host application can still be triggered.The latest research applies taint analysis and symbolic execution to transform the PoC of third-party libraries to make it suitable for host applications.However,there are often differences between the test environment of the third-party library and the real environment of the host application (they are heterogeneous environments),so that the PoC transformed by the above method is still difficult to apply to the host application.To solve the above problems,a method for PoC refactoring in heterogeneous environment is proposed,which can be divided into four steps.Firstly,we exeract the execution traces in the third-party library test environment and the host application environment respectively when the original PoC is input.Secondly,we compare and analyze the two traces obtained in the first step to identify differences.Thirdly,we analyze codes at difference points to identify the key variables that cause the diffe-rences.Finally,we locate the key fields in the PoC that can affect the state of key variables,by mutating the key fields of the PoC,we try to modify the state of the key variables and align the difference paths,guide the execution flow of the host application to reach the vulnerability code,and eventually we complete the refactoring of the PoC.Experiments are carried out on 11 real-world PoCs,and the experimental results show that the proposed method can successfully verify the triggerability of the propagated vu-lnerability in the host application in a heterogeneous environment.
Binary Code Similarity Detection Method Based on Pre-training Assembly Instruction Representation
WANG Taiyan, PAN Zulie, YU Lu, SONG Jingbin
Computer Science. 2023, 50 (4): 288-297.  doi:10.11896/jsjkx.220300271
Abstract PDF(4958KB) ( 369 )   
References | Related Articles | Metrics
Binary code similarity detection has been widely used in vulnerability searching,malware detection,advanced program analysis and other fields in recent years,while program code is similar to natural language in a degree,researchers start to use pre-training and other natural language processing related technologies to improve accuracy.A binary code similarity detection method based on pre-training assembly instruction representation is proposed to deal with the accuracy bottleneck due to insufficient consideration of instruction probability features.It includes tokenization method for multi-arch assembly instructions,and pre-trai-ning tasks that considering control flow,data flow,instruction logic and probability of occurrence,to achieve better vectorized representation of instructions.Downstream binary code similarity detection task is improved by combining pre-training method to gain accuracy boost.Experiments show that,compared with the existing methods,the proposed method improves instruction representing performance by 23.7% at the maximum,and improves block searching ability and similarity detection performance by up to 33.97% and 400% respectively.
Container-based Intrusion Detection Method for Cisco IOS-XE
YANG Pengfei, CAI Ruijie, GUO Shichen, LIU Shengli
Computer Science. 2023, 50 (4): 298-307.  doi:10.11896/jsjkx.220300264
Abstract PDF(2486KB) ( 312 )   
References | Related Articles | Metrics
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.
Anonymous Batch Authentication Scheme in Internet of Vehicles for WAVE Security Services
GUO Nan, SONG Xiaobo, ZHUANG Luyuan, ZHAO Cong
Computer Science. 2023, 50 (4): 308-316.  doi:10.11896/jsjkx.220300082
Abstract PDF(2346KB) ( 281 )   
References | Related Articles | Metrics
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.
Smart Contract Vulnerability Detection Based on Abstract Syntax Tree Pruning
LIU Zerun, ZHENG Hong, QIU Junjie
Computer Science. 2023, 50 (4): 317-322.  doi:10.11896/jsjkx.220300063
Abstract PDF(1819KB) ( 324 )   
References | Related Articles | Metrics
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.
Black-box Fuzzing Method Based on Reverse-engineering for Proprietary Industrial Control Protocol
YANG Yahui, MA Rongkuan, GENG Yangyang, WEI Qiang, JIA Yan
Computer Science. 2023, 50 (4): 323-332.  doi:10.11896/jsjkx.211200258
Abstract PDF(2582KB) ( 398 )   
References | Related Articles | Metrics
The wide application of industrial control proprietary protocols has brought great challenges to the safe operation of industrial control systems.Due to the closed-source nature of industrial control proprietary protocol specifications,it is difficult for traditional fuzzing testing tools to efficiently generate test cases,limiting the efficiency of fuzzing testing of industrial control equipment using proprietary industrial control protocols.A black box fuzzing method is proposed to solve this problem based on the reverse of a private industrial control protocol.First,an improved multiple sequence alignment algorithm and a field division algorithm are used to obtain the protocol field structure based on traffic capture.Then a series of heuristic rules are defined to identify the constant field,the serial number field,the length field,and the function code field in the protocol to infer the protocol format.After that,a protocol state machine is built according to the sequence and function code fields.In the process of fuzzing,according to the protocol format of reverse inference,various mutation strategies are used to generate test cases,and the constructed protocol state machine is used to guide the in-depth interaction between the fuzzing tool and the device under test.Based on the above methods,the ICPPfuzz tool is designed and implemented.The protocol reverse capability and fuzzing test capability of ICPPfuzz are evaluated with real equipment using three industrial control protocols(Modbus/TCP,UMAS,S7comm).Experimental results show that the tool’s field division,semantic recognition,and protocol state machine construction capabilities are significantly better than Netzob in protocol reversal.In terms of fuzzing test,the number of effective test cases generated by the tool within the same time is 1.25 times that of Boofuzz,and the quality of test cases and vulnerability discovery ability are also better than Boofuzz.At the same time,three denials of service vulnerabilities are successfully found when testing Modicon TM200/221 series PLC,which proves the tool’s effectiveness.
Multi-objective Federated Learning Evolutionary Algorithm Based on Improved NSGA-III
ZHONG Jialin, WU Yahui, DENG Su, ZHOU Haohao, MA Wubin
Computer Science. 2023, 50 (4): 333-342.  doi:10.11896/jsjkx.220300033
Abstract PDF(4251KB) ( 329 )   
References | Related Articles | Metrics
Federated learning technology solves the problems of data islands and privacy leakage to a certain extent.However it has shortcomings such as high communication cost,unstable communication,and uneven distribution of participant performance.In order to overcome these shortcomings and achieve a balance between model effectiveness,fairness,and communication costs,an improved NSGA-III algorithm for multi-objective optimization of federated learning is proposed.First,a federated learning multi-objective optimization model is constructed to maximize the accuracy of the global model,minimize the variance of the global mo-del accuracy distribution and minimize the communication cost of participant,and an improved NSGA-III algorithm based on fast greedy initialization is proposed,which improves the efficiency of NSGA-III for multi-objective optimization of federated learning.Experimental results show that the proposed optimization method can obtain a better Pareto solution than the classical multi-objective evolutionary algorithm.Compared with the standard model of federated learning,the optimized model can effectively lower the communication cost and the variance of the global model accuracy distribution while ensuring the accuracy of the global model.
Detection of Web Command Injection Vulnerability for Cisco IOS-XE
HE Jie, CAI Ruijie, YIN Xiaokang, LU Xuanting, LIU Shengli
Computer Science. 2023, 50 (4): 343-350.  doi:10.11896/jsjkx.220100113
Abstract PDF(2014KB) ( 418 )   
References | Related Articles | Metrics
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.
Android Malware Family Classification Method Based on Synthetic Image and Xception Improved Model
YU Xingzhan, LU Tianliang, DU Yanhui, WANG Xirui, YANG Cheng
Computer Science. 2023, 50 (4): 351-358.  doi:10.11896/jsjkx.220300200
Abstract PDF(2991KB) ( 271 )   
References | Related Articles | Metrics
Aiming at the problems in the field of Android malicious family detection,such as insufficient code visualization method construction information,large classification effect affected by the number of data sets and low classification accuracy,an Android malicious family classification method based on multi feature file synthetic image and Xception improved model is proposed.Fir-stly,three feature files corresponding to RGB multi-channel are selected to synthesize color images.Then,the improved Xception model introduces the focal loss function to alleviate the negative impact caused by the uneven distribution of samples.Finally,the attention mechanism is integrated into the improved model to extract the image features of malicious code from different dimensions,which improves the classification effect of the model.Experimental results show that the malicious code images synthesized by the proposed method contain richer features,have higher accuracy than the mainstream malicious family classification methods,and have better classification effect for unbalanced data sets.
Heterogeneous Provenance Graph Learning Model Based APT Detection
DONG Chengyu, LYU Mingqi, CHEN Tieming, ZHU Tiantian
Computer Science. 2023, 50 (4): 359-368.  doi:10.11896/jsjkx.220300040
Abstract PDF(3690KB) ( 391 )   
References | Related Articles | Metrics
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.
Interdiscipline & Frontier
Key Technologies of Intelligent Identification of Biomarkers:Review of Research on Association Prediction Between Circular RNA and Disease
HU Xuegang, LI Yang, WANG Lei, LI Peipei, YOU Zhuhong
Computer Science. 2023, 50 (4): 369-387.  doi:10.11896/jsjkx.220500114
Abstract PDF(4129KB) ( 296 )   
References | Related Articles | Metrics
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.
WiDoor:Close-range Contactless Human Identification Approach
CAO Chenyang, YANG Xiaodong, DUAN Pengsong
Computer Science. 2023, 50 (4): 388-396.  doi:10.11896/jsjkx.220300278
Abstract PDF(5846KB) ( 248 )   
References | Related Articles | Metrics
The rapid development of contactless identification technology based on Wi-Fi sensing has shown excellent application potential in the fields of intelligent human-computer interaction and intelligent security.However,it has been found that in narrow indoor scenarios,accuracy of existing lightweight identification model decreases with the shortening of transceiver distance.To solve the above problem,a close-range and contactless identification method,WiDoor,is proposed.During the data acquisition stage,WiDoor optimizes antenna deployment at receiving end based on Fresnel propagation model,and reconstructs gait information of multiple antennas to obtain the more complete gait description.In the identification stage,a lightweight convolution model,which combines the concatenated convolution module and the multi-scale convolution module,is used to reduce computational complexity while ensuring high identification accuracy.Experimental results show that WiDoor achieves identification accuracy rate of 99.1% on the 10-person dataset collected at the transceiver distance of 1 m.Moreover,parameter quantity of identification model is only 2% of those with the same identification accuracy,which outperforms other similar methods.
Batched Eigenvalue Decomposition Algorithms for Hermitian Matrices on GPU
HUANG Rongfeng, LIU Shifang, ZHAO Yonghua
Computer Science. 2023, 50 (4): 397-403.  doi:10.11896/jsjkx.220100232
Abstract PDF(2144KB) ( 297 )   
References | Related Articles | Metrics
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.