About Publications Experience


photo

Xingshan Zeng (Hamson)

Xingshan Zeng (曾幸山)

zxshamson AT gmail.com / xszeng AT se.cuhk.edu.hk / Google Scholar / GitHub

I am currently a PhD student in Systems Engineering and Engineering Management (SEEM) Dapartment, The Chinese University of Hong Kong (CUHK). My supervisor is Prof. Kam-Fai Wong. I got my B.S. degree from the department of Computer Science and Technology, University of Science and Technology of China (USTC) in 2016.

My research interests include Natural Language Processing (NLP), Social Media Analysis and Recommender Systems.

I have served as program committee of ACL, EMNLP, AAAI, etc.

Publications

Microblog Conversation Recommendation via Joint Modeling of Topics and Discourse
Xingshan Zeng, Jing Li, Lu Wang, N. Beauchamp, S. Shugars, and Kam-Fai Wong. NAACL 2018 [PDF] [Slides] [Code]

We propose a statistical model that jointly captures: (1) topics for representing user interests and conversation content, and (2) discourse modes for describing user replying behavior and conversation dynamics, for conversation recommendation.

Joint Effects of Context and User History for Predicting Online Conversation Re-entries
Xingshan Zeng, Jing Li, Lu Wang, and Kam-Fai Wong. ACL 2019 [PDF] [Slides] [Code&Data] [arXiv]

We study re-entry prediction foreseeing whether a user will return to a conversation they once participated in. We propose a neural framework with three main layers, each modeling context, user history, and interactions between them, to explore how the conversation context and user chatting history jointly result in their re-entry behaviors.

Neural Conversation Recommendation with Online Interaction Modeling
Xingshan Zeng, Jing Li, Lu Wang, and Kam-Fai Wong. EMNLP 2019 [PDF] [Poster] [Code&Data]

We present a novel framework to automatically recommend conversations to users based on their prior conversation behaviors. Built on neural collaborative filtering, our model incorporates graph-structured networks, where both replying relations and temporal features are encoded as conversation context.

Dynamic Online Conversation Recommendation
Xingshan Zeng, Jing Li, Lu Wang, Zhiming Mao, and Kam-Fai Wong. ACL 2020 [PDF] [Code&Data]

We study dynamic online conversation recommendation, to help users engage in conversations that satisfy their evolving interests. We propose a neural model, which can capture the temporal aspects of user interests, and cater for cold start problem in conversation recommendation.

Experience

Professional Experience

2020: PC member of AAAI, ACL, etc.
2019: PC member of EMNLP; Sub-Reviewer of ACL, NAACL.
2018: PC member of EMNLP; Sub-Reviewer of ACL, NAACL.

Teaching Assistance Experience

Data Structures: 2017 Spring
Information Technology Management (MSc Course): 2019 Spring, 2020 Spring
Information Systems Analysis and Design: 2018 Fall, 2019 Fall
Engineering Innovation and Entrepreneurship: 2016 Fall, 2017 Fall, 2018 Fall, 2020 Spring