Xunxun Chen
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Xunxun Chen has a total of 102 co-authors. Here are the top co-authors by number of shared publications:
Yongzheng Zhang
Chinese Academy of Sciences,Institute of Information EngineeringBeijingChina
6 shared publicationsGaochao Li
Institute of Information Engineering, Chinese Academy of Sciences, School of Cyber Security, University of Chinese Academy of Sciences
4 shared publicationsWei Wang
The Forth People's Hospital of Foshan, Foshan, 528000, China
4 shared publicationsXiao Wu
National Computer Network, Emergency Response Technical, Team/Coordination Center of China, Beijing, China
4 shared publicationsLiang Chen
Centre for Tuberculosis Control of Guangdong Province, 485 West Huangpu Avenue, Guangzhou, 510630, China
3 shared publicationsShuhao Li
Institute of Information Engineering, Chinese Academy of Sciences; School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China
3 shared publications
Latest Publications
- Open Access
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (4) , 5276-5284
Deepfake Video Detection via Facial Action Dependencies Estimation
Deepfake video detection has drawn significant attention from researchers due to the security issues induced by deepfake videos. Unfortunately, most of the existing deepfake detection approaches have not competently modeled the natural structures and movements of human faces. In this paper, we formulate the deepfake video detection problem into a graph classification task, and propose a novel paradigm named Facial Action Dependencies Estimation (FADE) for deepfake video detection. We propose a Multi-Dependency Graph Module (MDGM) to capture abundant dependencies among facial action units, and extracts subtle clues in these dependencies. MDGM can be easily integrated into the existing frame-level detection schemes to provide significant performance gains. Extensive experiments demonstrate the superiority of our method against the state-of-the-art methods.
- 26 June 2023
- Published by Association for the Advancement of Artificial Intelligence (AAAI) in Proceedings of the AAAI Conference on Artificial Intelligence
- Vol. 37 (4) , 5276-5284
Open AccessDeepfake Video Detection via Facial Action Dependencies Estimation
Deepfake video detection has drawn significant attention from researchers due to the security issues induced by deepfake videos. Unfortunately, most of the existing deepfake detection approaches have not competently modeled the natural structures and movements of human faces. In this paper, we formulate the deepfake video detection problem into a graph classification task, and propose a novel paradigm named Facial Action Dependencies Estimation (FADE) for deepfake video detection. We propose a Multi-Dependency Graph Module (MDGM) to capture abundant dependencies among facial action units, and extracts subtle clues in these dependencies. MDGM can be easily integrated into the existing frame-level detection schemes to provide significant performance gains. Extensive experiments demonstrate the superiority of our method against the state-of-the-art methods.
- 11 March 2023
- Published by Springer Nature
- p. 467-484
D$^{{3}}$: Duplicate Detection Decontaminator for Multi-Athlete Tracking in Sports Videos- 11 March 2023
- Published by Springer Nature
- p. 467-484
D$^{{3}}$: Duplicate Detection Decontaminator for Multi-Athlete Tracking in Sports Videos- 22 October 2020
- Published by Elsevier in Computers & Security
- Vol. 102, 102085
MASA: An efficient framework for anomaly detection in multi-attributed networks
- 22 October 2020
- Published by Elsevier in Computers & Security
- Vol. 102, 102085
MASA: An efficient framework for anomaly detection in multi-attributed networks
- 7 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. PP (15455971) , 1
Structured Sparsity Model Based Trajectory Tracking Using Private Location Data Release
Mobile devices have been an integral part of our everyday lives. Users' increasing interaction with mobile devices brings in significant concerns on various types of potential privacy leakage, among which location privacy draws the most attention. Specifically, mobile users' trajectories constructed by location data may be captured by adversaries to infer sensitive information. In previous studies, differential privacy has been utilized to protect published trajectory data with rigorous privacy guarantee. Strong protection provided by differential privacy distorts the original locations or trajectories using stochastic noise to avoid privacy leakage. In this paper, we propose a novel location inference attack framework, iTracker, which simultaneously recovers multiple trajectories from differentially private trajectory data using the structured sparsity model. Compared with the traditional recovery methods based on single trajectory prediction, iTracker, which takes advantage of the correlation among trajectories discovered by the structured sparsity model, is more effective in recovering multiple private trajectories simultaneously. iTracker successfully attacks the existing privacy protection mechanisms based on differential privacy. We theoretically demonstrate the near-linear runtime of iTracker, and the experimental results using two real-world datasets show that iTracker outperforms existing recovery algorithms in recovering multiple trajectories.
- 7 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. PP (15455971) , 1
Structured Sparsity Model Based Trajectory Tracking Using Private Location Data Release
Mobile devices have been an integral part of our everyday lives. Users' increasing interaction with mobile devices brings in significant concerns on various types of potential privacy leakage, among which location privacy draws the most attention. Specifically, mobile users' trajectories constructed by location data may be captured by adversaries to infer sensitive information. In previous studies, differential privacy has been utilized to protect published trajectory data with rigorous privacy guarantee. Strong protection provided by differential privacy distorts the original locations or trajectories using stochastic noise to avoid privacy leakage. In this paper, we propose a novel location inference attack framework, iTracker, which simultaneously recovers multiple trajectories from differentially private trajectory data using the structured sparsity model. Compared with the traditional recovery methods based on single trajectory prediction, iTracker, which takes advantage of the correlation among trajectories discovered by the structured sparsity model, is more effective in recovering multiple private trajectories simultaneously. iTracker successfully attacks the existing privacy protection mechanisms based on differential privacy. We theoretically demonstrate the near-linear runtime of iTracker, and the experimental results using two real-world datasets show that iTracker outperforms existing recovery algorithms in recovering multiple trajectories.
- Open Access
- 9 January 2020
- Published by Springer Nature in Journal of Molecular Evolution
- Vol. 88 (2) , 202-209
Potential Genes Related to Levofloxacin Resistance in Mycobacterium tuberculosis Based on Transcriptome and Methylome Overlap Analysis
Drug-resistant Mycobacterium tuberculosis (M. tuberculosis) has become an increasingly serious public health problem and has complicated tuberculosis (TB) treatment. Levofloxacin (LOF) is an ideal anti-tuberculosis drug in clinical applications. However, the detailed molecular mechanisms of LOF-resistant M. tuberculosis in TB treatment have not been revealed. Our study performed transcriptome and methylome sequencing to investigate the potential biological characteristics of LOF resistance in M. tuberculosis H37Rv. In the transcriptome analysis, 953 differentially expressed genes (DEGs) were identified; 514 and 439 DEGs were significantly downregulated and upregulated in the LOF-resistant group and control group, respectively. The KEGG pathway analysis revealed that 97 pathways were enriched in this study. In the methylome analysis, 239 differentially methylated genes (DMGs) were identified; 150 and 89 DMGs were hypomethylated and hypermethylated in the LOF-resistant group and control group, respectively. The KEGG pathway analysis revealed that 74 pathways were enriched in this study. The overlap study suggested that 25 genes were obtained. It was notable that nine genes expressed downregulated mRNA and upregulated methylated levels, including pgi, fadE4, php, cyp132, pckA, rpmB1, pfkB, acg, and ctpF, especially cyp132, pckA, and pfkB, which were vital in LOF-resistant M. tuberculosis H37Rv. The overlapping genes between transcriptome and methylome could be essential for studying the molecular mechanisms of LOF-resistant M. tuberculosis H37Rv. These results may provide informative evidence for TB treatment with LOF.
- 9 January 2020
- Published by Springer Nature in Journal of Molecular Evolution
- Vol. 88 (2) , 202-209
Open AccessPotential Genes Related to Levofloxacin Resistance in Mycobacterium tuberculosis Based on Transcriptome and Methylome Overlap Analysis
Drug-resistant Mycobacterium tuberculosis (M. tuberculosis) has become an increasingly serious public health problem and has complicated tuberculosis (TB) treatment. Levofloxacin (LOF) is an ideal anti-tuberculosis drug in clinical applications. However, the detailed molecular mechanisms of LOF-resistant M. tuberculosis in TB treatment have not been revealed. Our study performed transcriptome and methylome sequencing to investigate the potential biological characteristics of LOF resistance in M. tuberculosis H37Rv. In the transcriptome analysis, 953 differentially expressed genes (DEGs) were identified; 514 and 439 DEGs were significantly downregulated and upregulated in the LOF-resistant group and control group, respectively. The KEGG pathway analysis revealed that 97 pathways were enriched in this study. In the methylome analysis, 239 differentially methylated genes (DMGs) were identified; 150 and 89 DMGs were hypomethylated and hypermethylated in the LOF-resistant group and control group, respectively. The KEGG pathway analysis revealed that 74 pathways were enriched in this study. The overlap study suggested that 25 genes were obtained. It was notable that nine genes expressed downregulated mRNA and upregulated methylated levels, including pgi, fadE4, php, cyp132, pckA, rpmB1, pfkB, acg, and ctpF, especially cyp132, pckA, and pfkB, which were vital in LOF-resistant M. tuberculosis H37Rv. The overlapping genes between transcriptome and methylome could be essential for studying the molecular mechanisms of LOF-resistant M. tuberculosis H37Rv. These results may provide informative evidence for TB treatment with LOF.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
A Linguistics-based Stacking Approach to Disposable Domains Detection
More Internet services tend to collect the one-time information from clients via DNS queries. Notably, the uncertainty of such transient information makes these domain names be queried only once in their lifetime. This type of domain is called disposable domain. Although they are not malicious, the efficiency of DNS infrastructures will still be affected by their ever-increasing number. In this paper, we propose Vogers, a linguistics-based stacking model, to detect the disposable domains. Our evaluation demonstrates that Vogers decreases the false positive rate by more than 19%, compared with the prior art, while maintaining the true positive rate above 98.9%
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
A Linguistics-based Stacking Approach to Disposable Domains Detection
More Internet services tend to collect the one-time information from clients via DNS queries. Notably, the uncertainty of such transient information makes these domain names be queried only once in their lifetime. This type of domain is called disposable domain. Although they are not malicious, the efficiency of DNS infrastructures will still be affected by their ever-increasing number. In this paper, we propose Vogers, a linguistics-based stacking model, to detect the disposable domains. Our evaluation demonstrates that Vogers decreases the false positive rate by more than 19%, compared with the prior art, while maintaining the true positive rate above 98.9%
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 232-239
MURITE-Detector: Identifying User-Role in Information Theft Events of Mobile Network
The emergence of information theft apps poses a serious threat to smartphone users. Most of information theft apps rely on network interfaces to steal users' privacy and use short message service (SMS) to implement command and control. In this paper, we propose an available and effective user-role identification model, MURITE-detector (Mobile User-Role in Information Theft Events detector), by using network flows and call detail records (CDRs) with convolutional neural network (CNN) algorithm. Firstly, we generate network flow vectors and CDR vectors from raw data sets, and then match them into node vectors. Subsequently, we use CNN to classify user-roles into: Sourcer, Transferer, Victim and Other. Because of command-and-control server invalidation and system version incompatibility, etc., most of the collected information theft apps can't run properly in reality. So we extract code modules from some of these apps, and then recode and compile them into ITM-capsule (Information Theft Modules capsule) to generate information theft network traffic. Finally, we obtain 37,384 information theft network flows, 61,635 benign network flows and 200,522 short message CDRs. We match these data through labels and construct two node vector sets A and B randomly. In addition, we also compare CNN with other machine learning algorithms, and the result shows that CNN performs better. In an evaluation of MURITE-detector, it gets an accuracy of 92.17%, a precision of 93.18% and a recall of 94.68%. Therefore, our model is suitable for identifying user-role in mobile network information theft events.
- 1 August 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 232-239
MURITE-Detector: Identifying User-Role in Information Theft Events of Mobile Network
The emergence of information theft apps poses a serious threat to smartphone users. Most of information theft apps rely on network interfaces to steal users' privacy and use short message service (SMS) to implement command and control. In this paper, we propose an available and effective user-role identification model, MURITE-detector (Mobile User-Role in Information Theft Events detector), by using network flows and call detail records (CDRs) with convolutional neural network (CNN) algorithm. Firstly, we generate network flow vectors and CDR vectors from raw data sets, and then match them into node vectors. Subsequently, we use CNN to classify user-roles into: Sourcer, Transferer, Victim and Other. Because of command-and-control server invalidation and system version incompatibility, etc., most of the collected information theft apps can't run properly in reality. So we extract code modules from some of these apps, and then recode and compile them into ITM-capsule (Information Theft Modules capsule) to generate information theft network traffic. Finally, we obtain 37,384 information theft network flows, 61,635 benign network flows and 200,522 short message CDRs. We match these data through labels and construct two node vector sets A and B randomly. In addition, we also compare CNN with other machine learning algorithms, and the result shows that CNN performs better. In an evaluation of MURITE-detector, it gets an accuracy of 92.17%, a precision of 93.18% and a recall of 94.68%. Therefore, our model is suitable for identifying user-role in mobile network information theft events.
- 1 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
A Comprehensive Measurement Study of Domain-Squatting Abuse
Domain-squatting abuse refers to the premeditated attempt by an attacker to register perceptively confusing domain names thereby tricking visitors into querying them. There are totally five squatting types have been investigated so far, namely typo-squatting, bit-squatting, homograph-squatting, sound-squatting, and combo-squatting. Existing researches only focus on one specific squatting type and never explore the relationship among them. In this paper, we perform the first comprehensive measurement study of domain-squatting abuse. We select 786 the most queried domains, and hunt for squatting abuses against them in ISP-level DNS traffic. We find that although typo-squatting accounts for most of squatting domains, combo-squatting are able to attract more traffic. Our further case studies show that parking ads is still the most important way for attackers to make profits. The only exception is combo-squatting, in which squatters tend to leverage the reputation of squatted domains to develop their own business. It is worth noting that some squatting domains are even used to deliver malware. Moreover, the Alexa ranks of certain squatting domains have already surpassed the original domains. These results clearly call for the need to better protect the intellectual property of domain names.
- 1 May 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-6
A Comprehensive Measurement Study of Domain-Squatting Abuse
Domain-squatting abuse refers to the premeditated attempt by an attacker to register perceptively confusing domain names thereby tricking visitors into querying them. There are totally five squatting types have been investigated so far, namely typo-squatting, bit-squatting, homograph-squatting, sound-squatting, and combo-squatting. Existing researches only focus on one specific squatting type and never explore the relationship among them. In this paper, we perform the first comprehensive measurement study of domain-squatting abuse. We select 786 the most queried domains, and hunt for squatting abuses against them in ISP-level DNS traffic. We find that although typo-squatting accounts for most of squatting domains, combo-squatting are able to attract more traffic. Our further case studies show that parking ads is still the most important way for attackers to make profits. The only exception is combo-squatting, in which squatters tend to leverage the reputation of squatted domains to develop their own business. It is worth noting that some squatting domains are even used to deliver malware. Moreover, the Alexa ranks of certain squatting domains have already surpassed the original domains. These results clearly call for the need to better protect the intellectual property of domain names.
- Open Access
- 16 March 2019
- Published by Association for Computing Machinery (ACM)
- p. 54-59
A Deep Learning Based Fast-Flux and CDN Domain Names Recognition Method
- 16 March 2019
- Published by Association for Computing Machinery (ACM)
- p. 54-59
Open AccessA Deep Learning Based Fast-Flux and CDN Domain Names Recognition Method