janjanwang@gmail.com wangzhan@eduhk.hk

Postdoctoral Fellow

Analytics\Assessment Research Centre

The Education University of Hong Kong

Google scholar: https://scholar.google.com/citations?user=-D8lRSoAAAAJ&hl=zh-TW&oi=sra

ResearchGate: https://www.researchgate.net/profile/Zhan-Wang-25

Github: https://github.com/janjanwang

PhD supervisor: Peter Skehan (The tallest in the picture!)

Natural Language Processing (NLP) and Large Language Models (LLM) Applications in Education

Technology-supported language assessment, Generative AI in language education, Fake news detection, discourse and digital humanity

2. Improve Student Writing (ISW): ISW is a Multi-layered Feedback AI tool designed for offering structured, multi-level feedback for guiding second language learners’ argumentative essay (Wang et al., in preparation; Chiu et al., in preparation).

3. Conversational Agreement and Rapport Analytics (CARA): A natural language processing (NLP) tool for evaluating agreement and rapport in conversations (e.g., IELTS speaking, Chiu et al., 2025; Lin et al., 2025)

4. A multi-agent system for multi-lingual fake news detection: A persona-based agent system of LLMs that emulate human judgement of fake news in four languages (English, French, Korean, and Chinese) of tweets (Chiu et al., 2025).

What else I can do (to help colleagues and collaborators)?

While current technologies like machine learning and large language models (LLMs) have addressed bottom-up elements of language understanding—such as part-of-speech tagging at the word and sentence levels—higher-levels of human language and cognition, including discourse, context, metaphor, and sociolinguistics, remain at the forefront of technological innovation. Understanding human’s higher-level thinking requires a theoretical understanding of discourse and engineering approaches. As an interdisciplinary researcher, I aim to combine my linguistic and technological background to tackle these grand challenges.

Wang, Z., & Chiu, M. M. (2024). Multi-discourse modes in student writing: Effects of combining narrative and argument discourse modes on argumentative essay scores. Applied Linguistics. https://doi.org/10.1093/applin/amac073

Working with Hong Kong Examinations and Assessment Authority (HKEAA) to analyze HKDSE (a university entrance examination) Chinese paper, we sampled student writing on the following two argumentative task prompts:

Write an essay entitled ‘Not to be the first and not to be the last either’ to express your opinion of this attitude toward lives

Write an essay entitled ‘The sun and the shadow’ according to the pictures

N = 695 argumentative essays

1) What makes a good student argumentative writing? (We have official ratings from HKEAA)

2) Are there task differences? (fairness)

Wen, Y., Chiu, M., Guo, X., & Wang, Z. (2024). AI‐powered vocabulary learning for lower primary school students. British Journal of Educational Technology.

Currently teaching (Department of Chinese Language Studies, EdUHK)

Taught before (Department of English, The Chinese University of Hong Kong)

Interested in servicing as a faculty consultant for