A5078198240- Software-Defined Networks and 5G
- Network Security and Intrusion Detection
- Caching and Content Delivery
- Advanced Malware Detection Techniques
- Network Traffic and Congestion Control
- Internet Traffic Analysis and Secure E-voting
- Advanced Optical Network Technologies
- Cloud Computing and Resource Management
- Interconnection Networks and Systems
- Network Packet Processing and Optimization
- Adversarial Robustness in Machine Learning
- Anomaly Detection Techniques and Applications
- Software System Performance and Reliability
- Energy Efficient Wireless Sensor Networks
- Advanced Memory and Neural Computing
- Security and Verification in Computing
- Mobile Agent-Based Network Management
- Opportunistic and Delay-Tolerant Networks
- IoT and Edge/Fog Computing
- Spam and Phishing Detection
- Software Testing and Debugging Techniques
- Advanced Algorithms and Applications
- Software Engineering Research
- Advanced Computational Techniques and Applications
- Peer-to-Peer Network Technologies
Zhejiang University
2016-2025
Peng Cheng Laboratory
2024-2025
Zhejiang University of Science and Technology
2015-2024
Zhejiang Chinese Medical University
2024
Northeast Electric Power University
2009-2024
Wenzhou Medical University
2024
Taiyuan University of Technology
2024
China Jiliang University
2024
Electric Power University
2024
Jiangsu University of Science and Technology
2023
Online review system enables users to submit reviews about the products. However, openness of Internet and monetary rewards for crowdsourcing tasks stimulate a large number fraudulent write fake post advertisements interfere rank apps. Existing methods detecting spam have been successful but they usually aims at e-commerce (e.g. Amazon, eBay) recommendation Yelp, Dianping) systems. Since behaviors are complexity varying across different platforms, existing not suitable fraudster detection in...
Deep learning (DL) models are inherently vulnerable to adversarial examples - maliciously crafted inputs trigger target DL misbehave which significantly hinders the application of in security-sensitive domains. Intensive research on has led an arms race between adversaries and defenders. Such plethora emerging attacks defenses raise many questions: Which more evasive, preprocessing-proof, or transferable? effective, utility-preserving, general? Are ensembles multiple robust than individuals?...
Federated Learning is an effective solution to address the issues of data isolation and privacy leakage in machine learning. However, ensuring security network devices architectures deploying federated learning remains a challenge due attacks. This paper proposes attention-based Graph Neural Network for detecting cross-level cross-department method enables collaborative model training while protecting on distributed devices. By organizing traffic information chronological order constructing...
As the first defensive layer that attacks would hit, web application firewall (WAF) plays an indispensable role in defending against malicious like SQL injection (SQLi). With development of cloud computing, WAF-as-a-service, as one kind Security-as-a-service, has been proposed to facilitate deployment, configuration, and update WAFs cloud. Despite its tremendous popularity, security vulnerabilities WAF-as-a-service are still largely unknown, which is highly concerning given massive usage. In...
While the celebrated graph neural networks (GNNs) yield effective representations for individual nodes of a graph, there has been relatively less success in extending to task similarity learning. Recent work on learning considered either global-level graph–graph interactions or low-level node–node interactions, however, ignoring rich cross-level (e.g., between each node one and other whole graph). In this article, we propose multilevel matching network (MGMN) framework computing any pair...
Code retrieval is to find the code snippet from a large corpus of source repositories that highly matches query natural language description. Recent work mainly uses processing techniques process both texts (i.e., human language) and snippets machine programming language), however neglecting deep structured features codes, which contain rich semantic information. In this paper, we propose an end-to-end graph matching searching (DGMS) model based on neural networks for task retrieval. To end,...
With the growth of network applications such as 5G and artificial intelligence, security techniques, i.e., techniques that detect various attacks (e.g., well-known denial-of service (DDoS) attacks) prevent production networks data center networks) from being attacked, become increasingly essential for management have gained great popularity in networking community. Generally, these are built on proprietary hardware appliances, middleboxes, or paradigm combines both software-defined (SDN)...
A software-defined network (SDN) is increasingly deployed in many practical settings, bringing new security risks, e.g., SDN controller and switch hijacking. In this paper, we propose a real-time method to detect compromised devices reliable way. The proposed aims at solving the detection problem of when both are trustless, it complementary with existing methods. Our primary idea employ backup controllers audit handling information update events collected from its switches, by recognizing...
Fuzzing is a technique of finding bugs by executing target program recurrently with large number abnormal inputs. Most the coverage-based fuzzers consider all parts equally and pay too much attention to how improve code coverage. It inefficient as vulnerable only takes tiny fraction entire code. In this article, we design implement an evolutionary fuzzing framework called V-Fuzz, which aims find efficiently quickly in limited time for binary programs. V-Fuzz consists two main components: 1)...
Fuzzing is a technique of finding bugs by executing software recurrently with large number abnormal inputs. Most the existing fuzzers consider all parts equally, and pay too much attention on how to improve code coverage. It inefficient as vulnerable only takes tiny fraction entire code. In this paper, we design implement vulnerability-oriented evolutionary fuzzing prototype named V-Fuzz, which aims find efficiently quickly in limited time. V-Fuzz consists two main components: neural...
In this paper, we propose a novel Patch&Pair Convolutional Neural Networks (PPCNN) to distinguish Deepfake videos or images from real ones. Through the comprehensive evaluations on public datasets, demonstrate that our model performs better than existing detection methods and show generalization.
Given a large graph with millions of vertices and different types, how can we spot anomalies find potential adversaries in time? Most graph-based fraud detection algorithms focus on finding dense blocks, discovering local subgraphs, designing belief propagation, latent factor models. However, as fraudsters online social networks gradually become adversarial, distributed, invisible, it is easy for them to evade traditional methods. Even worse, existing systems are fragile the new attacks....
The alert correlation process that aggregates computer network security alerts to the same attack scenario provides a coherent view of status at higher abstraction level. This letter proposes framework called Alert-GCN correlate belong using graph convolutional networks (GCN). intuition is stacked layers help aggregate information from farther neighbors in graph, thus facilitating discovery. first transforms into with one-hot encoding and then feeds GCN perform node classification....
In network function virtualization (NFV), functions (NFs) are chained as a service chain (SFC) to enhance NF management with high flexibility. Recent solutions indicate that the processing performance of SFCs can be significantly improved by offloading NFs programmable switches. However, such requires deep understanding properties achieve maximum SFC performance, which brings non-trivial burdens administrators. this paper, we propose LightNF, novel system simplifies in networks. LightNF...
Aiming at defects such as low contrast in infrared ship images, uneven distribution of size, and lack texture details, which will lead to unmanned leakage misdetection slow detection, this paper proposes an detection model based on the improved YOLOv8 algorithm (R_YOLO).The incorporates Efficient Multi-Scale Attention mechanism (EMA), efficient Reparameterized Generalized-feature extraction module (CSPStage), small target header, Repulsion Loss function, context aggregation block (CABlock),...
Deep neural networks have been widely applied to bearing fault diagnosis systems and achieved impressive success recently.To address the problem that insufficient feature extraction ability of traditional methods results in poor effect under variable load noise interference scenarios, a rolling model combining Multi-Scale Convolutional Neural Network (MSCNN) Long Short-Term Memory (LSTM) fused with attention mechanism is proposed.To adaptively extract essential spatial information various...