Topological information is very important for understanding different types of online web services,in particular,for online social networks(OSNs).People leverage such information for various applications,such as socia...Topological information is very important for understanding different types of online web services,in particular,for online social networks(OSNs).People leverage such information for various applications,such as social relationship modeling,community detection,user profiling,and user behavior prediction.However,the leak of such information will also pose severe challenges for user privacy preserving due to its usefulness in characterizing users.Large-scale web crawling-based information probing is a representative way for obtaining topological information of online web services.In this paper,we explore how to defend against topological information probing for online web services,with a particular focus on online decentralized web services such as Mastodon.Different from traditional centralized web services,the federated nature of decentralized web services makes the identification of distributed crawlers even more difficult.We analyze the behavioral differences between legitimate users and crawlers in decentralized web services and highlight two key behavioral attributes that distinguish crawlers from legitimate users:instance interaction preferences and hop count in profile viewing patterns.Based on these insights:we propose a supervised machine learning-based framework for crawler detection,which is able to learn the federation-aware feature representations for users.To validate the framework’s effectiveness,we construct a labeled dataset that integrates real users with real-trace driven simulated crawlers in Mastodon.We use this dataset to train various supervised classifiers for crawler detection.Experimental results demonstrate that our framework can achieve an excellent classification performance.Moreover,it is observed that federation-aware features are effective in improving detection performance.展开更多
Network topology obfuscation is a technique aimed at protecting critical nodes and links from disruptions such as Link Flooding Attack(LFA).Currently,there are limited topology obfuscation methods for protecting criti...Network topology obfuscation is a technique aimed at protecting critical nodes and links from disruptions such as Link Flooding Attack(LFA).Currently,there are limited topology obfuscation methods for protecting critical nodes,and the existing approaches mainly achieve obfuscation by extensivelymodifying network links,resulting in high costs.To address this issue,this paper proposes a low-cost network topology obfuscation method dedicated to critical node protection,with its core innovation lying in a lightweight obfuscation architecture based on Fake Node Clusters(FNCs).Firstly,the protected network is modeled as an undirected graph,and an adjacency matrix is constructed to quantify the network scale and structural characteristics.Then,a fake node cluster generation algorithm is designed to construct an FNC adapted to the target network.Finally,a heuristic obfuscated topology generation algorithm is proposed.By optimizing the deployment positions of Fake Nodes Clusters(FNCs)in the protected network,this algorithm effectively reduces the number of FNCs required to generate the obfuscated topology,further lowering the obfuscation cost.Extensive experiments were conducted on the public Topology Zoo dataset,categorizing network topologies by node count into small-scale([0,50)),medium-scale([50,100)),and large-scale([100,200))groups.The experimental results demonstrate that the proposed approach achieves excellent obfuscation performance,reducing the critical node recognition rate to 0%.Compared to the typical method,EigenObfu,the proposed approach also reduces obfuscation costs by an average of 97.9%,99.6%,and 99.3%for small,medium,and large-scale networks,respectively.展开更多
基金funded by the National Key R&D Program of China under Grant(No.2022YFB3102901)National Natural Science Foundation of China(No.62072115,No.62102094)Shanghai Science and Technology Innovation Action Plan Project(No.22510713600).
文摘Topological information is very important for understanding different types of online web services,in particular,for online social networks(OSNs).People leverage such information for various applications,such as social relationship modeling,community detection,user profiling,and user behavior prediction.However,the leak of such information will also pose severe challenges for user privacy preserving due to its usefulness in characterizing users.Large-scale web crawling-based information probing is a representative way for obtaining topological information of online web services.In this paper,we explore how to defend against topological information probing for online web services,with a particular focus on online decentralized web services such as Mastodon.Different from traditional centralized web services,the federated nature of decentralized web services makes the identification of distributed crawlers even more difficult.We analyze the behavioral differences between legitimate users and crawlers in decentralized web services and highlight two key behavioral attributes that distinguish crawlers from legitimate users:instance interaction preferences and hop count in profile viewing patterns.Based on these insights:we propose a supervised machine learning-based framework for crawler detection,which is able to learn the federation-aware feature representations for users.To validate the framework’s effectiveness,we construct a labeled dataset that integrates real users with real-trace driven simulated crawlers in Mastodon.We use this dataset to train various supervised classifiers for crawler detection.Experimental results demonstrate that our framework can achieve an excellent classification performance.Moreover,it is observed that federation-aware features are effective in improving detection performance.
基金funded by the National Key Research and Development Program of China(Grant No.2022YFB3102900)the Key Science and Technology Project of Henan Province,China(No.252102211091).
文摘Network topology obfuscation is a technique aimed at protecting critical nodes and links from disruptions such as Link Flooding Attack(LFA).Currently,there are limited topology obfuscation methods for protecting critical nodes,and the existing approaches mainly achieve obfuscation by extensivelymodifying network links,resulting in high costs.To address this issue,this paper proposes a low-cost network topology obfuscation method dedicated to critical node protection,with its core innovation lying in a lightweight obfuscation architecture based on Fake Node Clusters(FNCs).Firstly,the protected network is modeled as an undirected graph,and an adjacency matrix is constructed to quantify the network scale and structural characteristics.Then,a fake node cluster generation algorithm is designed to construct an FNC adapted to the target network.Finally,a heuristic obfuscated topology generation algorithm is proposed.By optimizing the deployment positions of Fake Nodes Clusters(FNCs)in the protected network,this algorithm effectively reduces the number of FNCs required to generate the obfuscated topology,further lowering the obfuscation cost.Extensive experiments were conducted on the public Topology Zoo dataset,categorizing network topologies by node count into small-scale([0,50)),medium-scale([50,100)),and large-scale([100,200))groups.The experimental results demonstrate that the proposed approach achieves excellent obfuscation performance,reducing the critical node recognition rate to 0%.Compared to the typical method,EigenObfu,the proposed approach also reduces obfuscation costs by an average of 97.9%,99.6%,and 99.3%for small,medium,and large-scale networks,respectively.
