A5049770994- Privacy-Preserving Technologies in Data
- Cryptography and Data Security
- Cloud Data Security Solutions
- Photonic and Optical Devices
- Advanced Antenna and Metasurface Technologies
- Network Security and Intrusion Detection
- Electromagnetic Scattering and Analysis
- Sparse and Compressive Sensing Techniques
- Quantum Mechanics and Non-Hermitian Physics
- Advanced Malware Detection Techniques
- Electromagnetic Simulation and Numerical Methods
- Antenna Design and Analysis
- Topic Modeling
- Face recognition and analysis
- Antenna Design and Optimization
- Machine Fault Diagnosis Techniques
- Vehicular Ad Hoc Networks (VANETs)
- Software Testing and Debugging Techniques
- Neuroscience and Neural Engineering
- Human Mobility and Location-Based Analysis
- Mechanical and Optical Resonators
- Advanced Fiber Laser Technologies
- Intelligent Tutoring Systems and Adaptive Learning
- Online Learning and Analytics
- Biometric Identification and Security
Xiamen University of Technology
2024-2025
Hunan University
2016-2025
Guangzhou University
2023-2024
Sun Yat-sen University
2021
Uber AI (United States)
2021
Illinois Institute of Technology
2020-2021
ACT
2019-2021
City University of Seattle
2021
Chongqing Jiaotong University
2020
Campbell Institute
2020
Decision tree classification has become a prevailing technique for online diagnosis services. By outsourcing computation intensive tasks to cloud server, cloud-assisted services are better ways cases that the storage and requirements exceed capability of medical institutions. With privacy concerns as well intellectual property protection issues, valuable classifier sensitive user data should be protected against server. In this paper, we identify work-flow We propose an efficient secure...
Searchable encryption provides an effective mechanism that achieves secure search over encrypted data. A popular application model of searchable is a data owner stores to server and the can effectively perform keyword-based according query trapdoor submitted by user, where owner's user's queries are kept secret in server. Recently, many encryptions have been proposed achieve better security performance, provide updatable feature ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML"...
A mutual correlation between trajectories of two users is very helpful to real-life applications such as product recommendation and social media. While providing tremendous benefits, the releasing correlated may leak sensitive relations, due potential links correlations relations. To best our knowledge, we take first step propose a mathematically rigorous n-body Laplace framework, satisfying "-differential privacy, which efficiently prevents relation inference through n-node users. The...
Constructing three-dimensional (3D) aligned nanofiber scaffolds is significant for the development of cardiac tissue engineering, which promising in field drug discovery and disease mechanism study. However, current scaffold preparation strategy, mainly includes manual assembly hybrid 3D printing, faces challenge integrated fabrication morphology-controllable nanofibers due to its cross-scale structural feature. In this research, a trench-guided electrospinning (ES) strategy was proposed...
Malware detection becomes mission critical as its threats spread from personal computers to industrial control systems. Modern malware generally equips with sophisticated anti-detection mechanisms such code-morphism, which allows the evolve into many variants and bypass traditional code feature based In this paper, we propose disassemble binary executables opcodes sequences, then convert images. By using convolutional neural network compare opcode images generated targets known sample codes,...
In the digital age, data breaches have become a significant threat to consumer privacy, prompting implementation of stringent protection regulations worldwide. This paper evaluates effectiveness two prominent regulatory frameworks, General Data Protection Regulation (GDPR) in European Union and California Consumer Privacy Act (CCPA) United States, safeguarding responding breaches. Through comparative analysis their key provisions, enforcement mechanisms, real-world impacts, study examines...
Prior work has shown entropy-based DDOS detection is suitable for detecting low-rate DOS. A key challenge in this approach to determine the most threshold attacks accurately. To address challenge, we propose an advanced (AEB) scheme, which divides variable rate into different fields and treats each field with methods. Compared prior approaches, our scheme significant better performance on distinguishing waves of legal traffic flash crowds from We validate effectiveness by conducting...
New privacy implications are induced to individuals and families because of the time-series data classification problem in Internet Things such as appliance classifications smart grid. To prevent adversary from inferring household used grid, a singular spectrum analysis (SSA) has been applied local differential (SSA-LDP). First, Fourier noise added via geometric sum which proved achieve Laplace distribution. Furthermore, we have that sanitized through SSA-LDP is ε -deferentially private for...
Malware detection becomes mission critical as its threats spread from personal computers to industrial control systems. Modern malware generally equips with sophisticated anti-detection mechanisms such code-morphism, which allows the evolve into many variants and bypass traditional code feature based In this paper, we propose disassemble binary executables opcodes sequences, then convert images. By comparing opcode images generated targets known sample codes, can detect if target contain of...
Question answering (QA) is a hot field of research in Natural Language Processing. A big challenge this to answer questions from knowledge-dependable domain. Since traditional QA hardly satisfies some situations, such as disease diagnosis, drug recommendation, etc . In recent years, researches focus on knowledge-based question (KBQA). However, there still exist problems KBQA, KBQA limited by range historical cases and takes too much human labor. To address the problems, paper, we propose an...
Secure search techniques over encrypted cloud data allow an authorized user to query files of interest by submitting keywords the server in a privacy-preserving manner. However, practice, returned results may be incorrect or incomplete dishonest environment. For example, intentionally omit some qualified save computational resources and communication overhead. Thus, well-functioning secure system should provide verification mechanism that allows verify results. In this paper, we design...
Existing indoor semantic recognition schemes are capable of discovering patterns through smartphone sensing; however, there is a lack device-free system. In this paper, we propose WiFiMap+, which first-ever automatical inference system using WiFi signals to recognize highlevel semantics from human activities and environments, where the high-level consist facilities environments. To characterize static environments dynamical separately with channel state information (CSI), novel two-stream...
In the big data era, with rapid development of location-based applications, GPS enabled devices and institutions, location correlation privacy raises more people's concern. Because adversaries may combine correlations their background knowledge to guess users' privacy, such should be protected preserve privacy. order deal disclosure problem, perturbation generalization have been proposed. However, most proposed approaches depend on syntactic models without rigorous guarantee. Furthermore,...
With the rise of more interactive assessments, such as simulation- and game-based assessment, process data are available to learn about students' cognitive processes well motivational aspects. Since can be complicated due interdependencies in time, our traditional psychometric models may not necessarily fit, we need look for additional ways analyze data. In this study, draw from a study on self-adapted test under different goal conditions (Arieli-Attali, 2016) use hidden Markov takers'...
The high-speed millimeter wave Wireless Hadamard image transmission technique is studied for the next-generation 5G wireless network and beyond. Cyclic transform has been applied to image, followed by sub-sampling compression. For regularly sub-sampled transform, Fourier spectrum method can be used reconstruct when it transmitted receiver. randomly Compressed Sensing used. compatible with Multiple Inputs Outputs in mmWave networks. Both software hardware implementations of realized through...