Dr. Hongchen Wu is currently an associate professor in the School of Information Science and Engineering at Shandong Normal University. He received his Ph.D. in Computer Science and Technology from Shandong University in June 2016, and from September 2013 to September 2015, he was a joint Ph.D. student at the University of California, Irvine.
Dr. Hongchen Wu's research lies in the areas of next-generation internet and information security, big data computing based on deep learning, and interdisciplinary artificial intelligence research, including fake news detection, multimodal fake content (text, image, voice) recognition, personalized recommendation, privacy protection, computer-assisted education, and behavioral analysis. He is a member of the China Computer Federation (CCF) and a committee member of the CCF Service Computing Technical Committee. He serves as a reviewer for core computer science journals such as the Journal of Network and Computer Applications, IEEE Transactions (Cybernetics, MM, SC, EC, NNLS), Information Sciences, Expert Systems with Applications, and Neurocomputing, as well as interdisciplinary journals including Computers & Education, Computers in Human Behavior, and Multivariant Behavior Research.
Project Name: Key Technologies for Multimodal Fake Information Detection Targeting Deepfakes（面向深度伪造的多模态虚假信息检测关键技术）
Research Areas: Key Technology; Deepfake; Multimodal; Fake Information; Online Conformity Behavior
Project Description: Research on multimodal fake information detection for deepfake is a critical issue that needs to be addressed urgently in online public opinion monitoring and secured big data prediction with early warning systems. This project relies basically on computer science techniques and integrates interdisciplinary technologies such as psychology and cyber-behavioral science. It first establishes multimodal data sample repositories of text, images, and speech from network traffic information, consistently extracting encoded features such as text syntax, image texture, and speech emotion, etc., and generating multimodal semantic vectors through constructed data-augmented multi-layered semantic information space and identified fabrication traces. Then, this project further employs deep learning models to perceive contradictory semantics in sudden traffic surges, to address the challenge posed by the iterative advancement of deepfake technology, which makes fake information harder to detect and emotionally manipulates users to accelerate its spread, called online conformity behavior. By segmenting macro- and micro-semantic environments and amplifying the distinction between positive and negative samples, the approach enhances the real-time identification of forged features. Finally, dynamic communities are detected to identify opinion leaders who spread negative emotions through influence calculation in the semantic environment, where epidemic models are applied to limit the level of dissemination and random graph decay mechanism is employed to reduce interaction weights, curbing the dynamic expansion of fake information. This project provides key technical support and theoretical foundations for the security framework construction of multimodal fake news detection and intelligent network behavior guidance.
Project Name: Research on User Experience Based Cross-Domain Recommendation and Privacy Disclosure Pattern Mining（基于用户体验的跨域推荐及隐私分享行为模式研究）
Research Areas: Domain specific prediction technology; User experience; Cross-domain recommendation; Privacy disclosure behavior; Factor analysis
Project Description: This research is focusing on user datasets from the cross-domain recommendation, whose characters are “enormous, scattered and fragmented”, and analyses the sensitiveness association between users’ trust to different websites and their personal traits. A cross-sites user service agent will be designed, and the machine learning techniques are modified for recognizing users’ privacy disclosure differences and for clustering them into groups. The privacy default settings are applied to raise users’ initial willingness of disclosure, and the Decider collector is constructed for making the privacy integration strategy, which would track users’ trust fluctuation as well as their feedbacks in privacy disclosure, and thus could gain users’ trust and raise users’ disclosure amount. The confirmatory factor analysis is applied to construct the users’ behavioral model which is consist of the correlations between the trust values and the personal traits of disclosure, and the privacy protection and collection strategies are determined for fitting users’ disclosure behavior model under the cross-domain recommendation hierarchy in the process of model optimization.
Project Name: 面向跨平台用户的隐私管理理论与方法研究（ZR2017BF019）
Research Areas: Privacy management; Cross platform users; Differences in user behavior; Factor analysis method; Personalized service
Project Description: This study focuses on the characteristics of "diverse, scattered, and multifactorial" cross platform user data. Machine learning algorithms are used to analyze the weight correlation between cross platform users' trust in different platforms and their own psychological characteristics. The optimal sharing parameters are studied, and a hierarchical experience mathematical model is constructed to demonstrate the mechanism of emotional contagion. Test the differences in user behavior before and after privacy collection, analyze the trend of emotional peaks, construct privacy sharing feature parameters and gradient characteristics, ensure high trust while effectively protecting users' privacy sharing volume. Based on confirmatory factor analysis, develop a leverage mechanism for potential risks and personalized service quality improvement, and study the dynamic interest trajectory and pros and cons strategies of cross platform users. Summarize the joint analysis method of cross platform data, optimize the hierarchical structure, and establish the best management system for collecting user privacy sharing.
Project Name: 基于互联网用户行为认证的在线支付欺诈识别关键技术研究（2019GGX101075）
Research Areas: New generation information technology; Big data technology
Project Description: With the rapid development of the new generation of information technology and the Internet, users pay online to obtain fast and convenient e-commerce settlement services, but also increase concerns about security. This study is based on user behavior authentication in the big data environment, and constructs a pattern recognition mechanism for online payment fraud risk. By confirming the core event nodes in the payment process, we can predict the potential payment objects and default transaction entries of users, track the entire chain track process of users' payment, extract the characteristics and emotional fluctuations of fraudulent users by correlation analysis, predict the probability of transaction fraud, and solve the security guarantee problem of dynamic risks in the Internet.
Email: wuhongchen@sdnu.edu.cn
Office: Room 402, Wenzong Building, Changqinghu Campus