Dr Yangjun ZhangPostdoctoral Research AssistantEmail: yangjun.zhang@qmul.ac.ukRoom Number: Engineering, ENG 109ProfileTeachingResearchPublicationsSupervisionProfileI am currently a Postdoctoral Research Assistant at the Queen Mary University of London. My current research is funded by UKRI through the AP4L project. I have graduated with a Ph.D. in computer science from Information Retrieval Lab, at the University of Amsterdam, under the supervision of Prof. Maarten de Rijke. My research interests lie in natural language processing, information retrieval, and medical information processing (gene, behavior, and brain fMRI image).TeachingPreviously I had teaching experience in Fairness, Accountability, Confidentiality and Transparency in AI, Information Retrieval and DevOps and Cloud-based Software courses in University of Amsterdam (lab session, seminar, or project).ResearchResearch Interests:My research interests are mainly the following areas: social network analysis; safety of dialogue systems; information retrieval; gene, drugs and brain aging using fMRI and behaviour studies. In my PhD, I focused on malevolent dialogue response detection and evaluation. A malevolent response is a kind of dialogue response that might contain offensive or objectionable content including hate, insult, threat, etc. In my work, we first analyze the malevolence problem of state-of-the-art generation models with malevolence detection models. We also propose a knowledge pre-selection mechanism to improve informativeness before analyzing sequence to sequence based generation models. Second, we introduce taxonomies, datasets, and methods for single-label dialogue malevolence detection. Third, we build datasets and methods for multi-label dialogue malevolence detection from a single-label training set. The taxonomy of multi-label dialogue malevolence detection is the same as single-label detection. Finally, we propose a human-machine collaborative evaluation framework for dialogue malevolence evaluation. Currently, I’m working on the AP4L project, which focuses on the online privacy and vulnerability challenges that people face when during life transitions.PublicationsPlease find list of my publications on Google Scholar: https://scholar.google.com/citations?user=stmImoAAAAASupervisionPreviously I supervised several master theses in data science and artificial intelligence in University of Amsterdam; I also help supervise PhD students during my postdoctoral position in University of Amsterdam.