A5085450781- Network Security and Intrusion Detection
- Internet Traffic Analysis and Secure E-voting
- Software-Defined Networks and 5G
- Advanced Graph Neural Networks
- Caching and Content Delivery
- Advanced Malware Detection Techniques
- Network Packet Processing and Optimization
- Spam and Phishing Detection
- Topic Modeling
- Software Testing and Debugging Techniques
- Software Engineering Research
- Opportunistic and Delay-Tolerant Networks
- Complex Network Analysis Techniques
- Software Reliability and Analysis Research
- Imbalanced Data Classification Techniques
- Mobile Ad Hoc Networks
- Software System Performance and Reliability
- Privacy-Preserving Technologies in Data
- Recommender Systems and Techniques
- Digital and Cyber Forensics
- Privacy, Security, and Data Protection
- Hate Speech and Cyberbullying Detection
- Age of Information Optimization
- Energy Efficient Wireless Sensor Networks
- Blockchain Technology Applications and Security
Shenzhen Institute of Information Technology
2016-2025
Tsinghua University
2011-2017
University Town of Shenzhen
2016-2017
Wuhan University of Science and Technology
2017
ORCID
2017
Xi'an Jiaotong University
2006
In the current mobile edge computing (MEC) system, user dynamics, diversity of applications, and heterogeneity services have made cooperative service caching task offloading decision increasingly important. Service a naturally hierarchical structure, thus, reinforcement learning (HRL) can be used to effectively alleviate dimensionality curse in it. However, traditional HRL algorithms are designed for short-term missions with sparse rewards, while existing proposed MEC lack delicate coupling...
IPv6 geolocation is necessary for many location-based Internet services. However, the accuracy of current methods including machine-learning-based or deep-learning-based location algorithms are unsatisfactory users. Strong geographic correlation observed measurement path features close to target IP, so previous focused more on stable paths in vicinity probe. Based this, this paper proposes a new algorithm, SubvectorS_Geo, which mainly divided into three steps: firstly, it filters...
Current Internet packet delivery only relies on packet's destination IP address and forwarding devices neglect the validation of source address, it makes attackers can leverage this flaw to launch attacks with forged so as meet their vicious purposes avoid be tracked. In order mitigate threat enhance accountability, many solutions have been proposed either from intra-domain or inter-domain aspects. However, most them faced some issues hard cope with, e.g., low filtering rates, high...
In wireless sensor networks, it is a typical threat to source privacy that an attacker performs backtracing strategy locate nodes by analyzing transmission paths. With the popularity of Internet Things in recent years, protection has attracted lot attentions. order mitigate this threat, many proposals show their merits. However, they fail get tradeoff between multipath and cost. paper, we propose constrained random routing mechanism, which can constantly change next-hop instead relative...
Along with the popularity of online social network (OSN), more and OSN users tend to create their accounts in different platforms. Under such circumstances, identifying same user among OSNs offers tremendous opportunities for many applications, as identification, migration patterns, influence estimation, expert finding media. Different from existing solutions which employ profile or structure alone, this paper, we proposed a novel joint solution named MapMe, takes both feature into account,...
Personal privacy is facing severe threats as social networks are sharing user data with advertisers, application developers, and mining researchers. Although these anonymized by removing personal information, such identity, nickname, or address information still could not be protected effectively. In order to arouse the attention of people from academia industry for protection, we propose a random forest method de-anonymize networks. First, convert network de-anonymization problem into...
IP spoofing has nowadays become a research focus, as it been bothering netizens since the emergence of Internet. Though many studies have made their contributions to prevention IP-spoofing, most excellent one is SAVI (Source Address Validation Improvement) proposal advocated by IETF, can prevent IP-spoofing from happening automatically binding key properties hosts in layer2 access subnet. Nevertheless, till now, only focuses on IPv6 stack and simple network scenarios. To best our knowledge,...
Since the facts in knowledge graph (KG) cannot be updated automatically over time, some have temporal conflicts. To discover and eliminate conflicts KG, this paper proposes a novel conflict resolution method based on KG embedding (named TeCre). Firstly, predicate relation timestamp information of time series are incorporated into entity–relation representation by leveraging (KGE) method. Then, taking account chronological sequence evolution TeCre constrains according to principles disjoint,...
Link prediction in knowledge graph is the task of utilizing existing relations to infer new so as build a more complete graph. The inferred plus original symmetry completion Previous research on link predication only focuses path or semantic-based features, which can hardly have full insight features between entities and may result certain ratio false inference results. To improve accuracy predication, we propose novel approach named Entity Prediction for Knowledge Graph (ELPKG), achieve...