Kaiqi Zhao

Kaiqi Zhao

Senior Lecturer

Office:
Newmarket Campus, Building 903, Level 4, Room 428
School of Computer Science
314-390 Khyber Pass Road, Newmarket, Auckland 1023
Email: kaiqi.zhao[at]auckland.ac.nz
     

I am currently a Senior Lecturer with the School of Computer Science at the University of Auckland and a member of the Machine Learning Group @ UoA. I have broad research interests in machine learning and data mining. Specifically, my recent research focuses on spatio-temporal data mining, knowledge graphs and recommender systems. I am interested in building machine learning models to address practical problems such as geo-topic mining, GPS mobility modeling, and user preference mining. 

Prospective graduate student: I am looking for self-motivated and hard-working Honors/Master/Ph.D. students to do exciting research in spatio-temporal data mining, text mining, and recommender system. Please email your CV if you are interested!

Biography

  • 2022 – now:  Senior Lecturer, School of Computer Science, University of Auckland.
  • 2019 – 2022:  Lecturer, School of Computer Science, University of Auckland.
  • 2018 – 2019: Research Fellow, Singtel Cognitive and Aritificial Intelligence Lab for Enterprise, NTU.
  • 2013 – 2018: Ph.D. in Computer Science, School of Computer Science & Engineering, Nanyang Technological University. Supervisor: Dr. Gao CONG.
  • 2011 – 2012: Intern, Web Search & Mining Group, Microsoft Research Asia. Mentor: Dr. Haixun Wang.
  • 2010 – 2013: M.Eng. in Computer Engineering, Department of Computer Science, Shanghai Jiao Tong University. Supervisor: Dr. Kenny Q. Zhu.
  • 2007 – 2009: B.A. in Japanese Language, School of Foreign Languages, Huazhong University of Science and Technology.
  • 2005 – 2009: B.Eng. in Software Engineering, School of Software Engineering, Huazhong University of Science and Technology.

Research

Spatio-temporal data mining

The prevalence of location positioning devices such as GPS have made huge amount of geo-tagged data available. My research on spatio-temporal data mining is to automatically uncover latent patterns from data associated with GPS coordinates over time. This research area links to various of applications including GPS trajectory mining, local event detection, location prediction and traffic prediction.

Knowledge-enhanced NLP

Understanding of events and facts in plain text is challenging because of its unstructured nature. In this research area, I focus on developing effective methods for understanding text with the help of large knowledge graphs. Specifically, we study the problems of structural representation learning of domain knowledge for document classifications, summarisations and predictions. Besides, we study the cutting-edge problems of retrieval augmented generation and multi-modal data analogical reasoning for temporal knowledge graphs.

Recommender systems

Recommender system is a common practice to address the information overload problem and is widely applied in many industries such as journalism, e-commerce system and location-based services. My approach is to use advanced machine learning techniques on interesting recommendation problems, such as location-based recommendations, session-based recommendation, long-short term user preference modeling.

Publication

  1. Periormer: Periodic Transformer for seasonal and irregularly sampled time series
    Xiaobin Ren, Kaiqi Zhao, Katerina Taskova, Patricia Riddle, Lianyan Li
    CIKM, 2024
  2. SKGSum: Structured Knowledge-Guiding Document Summarization
    Qiqi Wang, Ruofan Wang, Kaiqi Zhao, Robert Amor, Benjamin Liu, Jiamou Liu, Xianda Zheng, Zijian Huang
    ACL (findings), 2024
  3. TP-GCN: Continuous Dynamic Graph Neural Network for Graph Classification
    Jie Liu, Jiamou Liu, Kaiqi Zhao, Yanni Tang, Wu Chen
    ICDE, 2024
  4. A Graph-based Representation Framework for Trajectory Recovery via Spatiotemporal Interval-Informed Seq2Seq
    Yaya Zhao, Kaiqi Zhao, Zhiqian Chen, Yuanyuan Zhang, Yalei Du, Xiaoling Lu
    IJCAI, 2024
  5. Contrastive Learning for Signed Bipartite Graphs
    Zeyu Zhang, Jiamou Liu, Kaiqi Zhao, Song Yang, Xianda Zheng, Yifei Wang
    SIGIR, 2023
  6. DAMR: Dynamic Adjacency Matrix Representation Learning for Multivariate Time Series Imputation
    Xiaobin Ren, Kaiqi Zhao, Patricia Riddle, Katerina Taskova, Qingyi Pan, Lianyan Li
    SIGMOD, 2023
  7. WISK: A Workload-aware Learned Index for Spatial Keyword Queries
    Yufan Sheng, Xin Cao, Yixiang Fang, Kaiqi Zhao, Jianzhong Qi, Gao Cong, Wenjie Zhang
    SIGMOD, 2023
  8. RSGNN: A Model-agnostic Approach for Enhancing the Robustness of Signed Graph Neural Networks
    Zeyu Zhang, Jiamou Liu, Xianda Zheng, Yifei Wang, Pengqian Han, Wang Yupan, Kaiqi Zhao, Zijian Zhang
    The Web Conference (WWW), 2023
  9. USER: Unsupervised Structural Entropy-based Robust Graph Neural Network  
    Yifei Wang, Yupan Wang, Zeyu Zhang, Song Yang, Kaiqi Zhao, Jiamou Liu
    AAAI, 2023
  10. ROLE: Rotated Lorentzian Graph Embedding Model for Asymmetric Proximity
    Shanshan Feng, Lisi Chen, Kaiqi Zhao, Wei Wei, Xuemeng Song, Shuo Shang, Panos Kalnis, Ling Shao
    TKDE, 2022
  11. H-Diffu: Hyperbolic Representations for Information Diffusion Prediction
    Shanshan Feng, Kaiqi Zhao, Lanting Fang, Kaiyu Feng, Wei Wei, Xutao Li, Ling Shao
    TKDE, 2022
  12. D2GCLF: Document-to-Graph Classifier for Legal Document Classification [pdf]
    Qiqi Wang, Kaiqi Zhao, Robert Amor, Benjamin Liu, Ruofan Wang
    NAACL (findings), 2022
  13. GETNext: Trajectory Flow Map Enhanced Transformer for Next POI Recommendation [codes]
    Song Yang, Jiamou Liu, Kaiqi Zhao
    SIGIR, 2022
  14. Interconnected Neural Linear Contextual Bandits with UCB Exploration 
    Yang Chen, Miao Xie, Jiamou Liu, Kaiqi Zhao
    PAKDD, 2022
  15. A knowledge-enriched ensemble method for word embedding and multi-sense embedding 
    Lanting Fang, Yong Luo, Kaiyu Feng, Kaiqi Zhao, Aiqun Hu
    TKDE, 2022
  16. Space Meets Time: Local Spacetime Neural Network For Traffic Flow Forecasting [arxiv][codes]
    Song Yang, Jiamou Liu, Kaiqi Zhao
    ICDM, 2021
  17. Uniqueness Constraints on Property Graphs 
    Philipp Skavantzos, Kaiqi Zhao, Sebastian Link
    CAiSE, 2021
  18. Node2LV: Squared Lorentzian Representations for Node Proximity
    Shanshan Feng, Lisi Chen, Kaiqi Zhao, Wei Wei, Fan Li, Shuo Shang
    ICDE, 2021
  19. PGeoTopic: A Distributed Solution for Mining Geographical Topic Models
    Kaiqi Zhao, Gao Cong, Xiucheng Li
    TKDE, 2020
  20. Online Anomalous Trajectory Detection with Deep Generative Sequence Modeling[pdf][codes]
    Yiding Liu, Kaiqi Zhao, Gao Cong, Zhifeng Bao
    ICDE, 2020
  21. EdgeRec: Recommender System on Edge in Mobile Taobao
    Yu Gong, Ziwen Jiang, Yufei Feng, Binbin Hu, Kaiqi Zhao, Qingwen Liu, Wenwu Ou
    CIKM, 2020
  22. A Novel Model for Imbalanced Data Classification
    Jian Yin, Chunjing Gan, Kaiqi Zhao, Xuan Lin, Zhe Quan, Zhi-Jie Wang
    AAAI, 2020
  23. Knowledge-Enhanced Ensemble Learning for Word Embeddings[codes]
    Lanting Fang, Yong Luo, Kaiyu Feng, Kaiqi Zhao, Aiqun Hu
    WWW, 2019
  24. Exploring Market Competition over Topics in Spatio-Temporal Document Collections
    Kaiqi Zhao,  Gao Cong, Jin-Yao Chin, Rong Wen.
    VLDB Journal, 2019
  25. Efficient Similar Region Search with Deep Metric Learning [pdf]
    Yiding Liu, Kaiqi Zhao, Gao Cong.
    SIGKDD, 2018
  26. ANR: Aspect-based Neural Recommender [pdf]
    Jin-Yao Chin, Kaiqi Zhao, Shafiq Joty, Gao Cong.
    CIKM, 2018
  27. Deep Representation Learning for Trajectory Similarity Computation [pdf][codes]
    Xiucheng Li, Kaiqi Zhao, Gao Cong, Christian S. Jensen, Wei Wei.
    ICDE, 2018
  28. Biclustering: An application of Dual Topic Models 
    Daniel Rugeles, Kaiqi Zhao, Gao Cong, Manoranjan Dash, Shonali  Krishnaswamy.
    SDM, 2017
  29. Annotating Points of Interest with Geo-tagged Tweets [pdf]
    Kaiqi Zhao, Gao Cong, Aixin Sun.
    CIKM, 2016
  30. Towards Personalized Maps: Mining User Preferences from Geo-textual Data (Demo paper) [pdf]
    Kaiqi Zhao, Yiding Liu, Quan Yuan, Lisi Chen, Zhida Chen, Gao Cong.
    VLDB, 2016
  31. A System for Region Search and Exploration (Demo paper) [pdf]
    Kaiyu Feng, Kaiqi Zhao, Yiding Liu, Gao Cong.
    VLDB, 2016
  32. Topic Exploration in Spatio-Temporal Document Collections 
    Kaiqi Zhao, Lisi Chen, Gao Cong.
    SIGMOD, 2016
  33. Querying and mining geo-textual data for exploration: Challenges and opportunities
    Gao Cong, Kaiyu Feng, Kaiqi Zhao.
    ICDE Workshops, 2016
  34. Representing Verbs as Argument Concepts [pdf]
    Yu Gong, Kaiqi Zhao, Kenny Q. Zhu.
    AAAI, 2016
  35. Who, where, when and what: a nonparametric bayesian approach to context-aware recommendation and search for twitter users
    Quan Yuan, Gao Cong, Kaiqi Zhao, Zongyang Ma, Aixin Sun.
    TOIS, 2015
  36. SAR: A Sentiment-Aspect-Region Model for User Preference Analysis in Geo-tagged Reviews [pdf]
    Kaiqi Zhao, Gao Cong, Quan Yuan, Kenny Q. Zhu.
    ICDE, 2015
  37. Clustering Image Search Results by Entity Disambiguation
    Kaiqi Zhao, Zhiyuan Cai, Qingyu Sui, Enxun Wei, Kenny Q. Zhu.
    ECML/PKDD, 2014
  38. Wikification via link co-occurrence
    Zhiyuan Cai, Kaiqi Zhao, Kenny Q. Zhu, Haixun Wang.
    CIKM, 2013
  39. CISC: Clustered Image Search by Conceptualization (Demo paper)
    Kaiqi Zhao, Enxun Wei, Qingyu Sui, Kenny Q. Zhu, Eric Lo.
    EDBT, 2013

Students

I am working with the following outstanding students:

Current PhDs

  • Philipp Skavantzos, 2019 – present, co-supervised with Prof. Sebastian Link. Research topic: Graph database.
  • Yufan Sheng, 2021 – present, co-supervised with Dr. Xin Cao (UNSW) and Dr. Yixiang Fang (CUHK-SZ). Research topic: Machine Learning for Databases.
  • Qiqi Wang, 2021 – present, co-supervised with Prof. Robert Amor and Dr. Benjamin Liu. Research topic: Knowledge-enhanced AI for Law
  • Xiaobin Ren, 2021 – present, co-supervised with Dr. Patricia Riddle and Dr. Katerina Taskova. Research topic: Spatio-temporal data mining for multi-source environmental data.
  • Eric Zheng, 2021 – present (part-time), co-supervised with Prof. Gill Dobbie. Research topic: Deep Bayesian Learning for Recommender Systems
  • Xianda Zheng, 2022 – present, co-supervised with Dr. Jiamou Liu. Research topic: Graph Representation Learning
  • Ziyi Jiang, 2022 – present, co-supervised with Prof. Gill Dobbie. Research topic: Trajectory Data Mining
  • Yanni Tang, 2023 – present, co-supervised with Dr. Jiamou Liu. Research topic: Graph representation learning for intelligent softwares

Alumni

  • Song Yang (PhD), graduated in 2023, Employment: Meta

Teaching

  • Algorithms for Massive Data (COMPSCI753), Instructor, S2 2020-2022, 2024, UoA
  • Data-mining and Machine Learning (COMPSCI760), Instructor, S2 2019-2021, UoA
  • Algorithms and Data Structures (COMPSCI220), Instructor, S1 2021,2024, UoA-NEFU
  • Machine Learning (COMPSCI361), Instructor, S1 2022-2024, UoA-NEFU
  • Artificial Intelligence (COMPSCI367), Instructor, S1 2020, UoA-SWU
  • Machine Learning (COMPSCI361), Guest Lecturer, S2 2019, UoA
  • Introduction to Database (CZ2007), Teaching Assistant, Fall 2016, NTU
  • Database System Principles (CZ4031), Teaching Assistant, Fall 2015, NTU
  • Windows Internals (CS490), Teaching Assistant, Fall 2010, SJTU

Professional Services

Conference services:

  • Regular PC member of SIGIR, KDD, ICDE, CIKM, AAAI, IJCAI
  • Area chair of KDD 2024
  • Senior PC of CIKM 2024

Journal Reviewer:

  • TOIS, TKDE, TKDD, IEEE Trans. on Big Data, World Wide Web Journal.

Conference Organisation:

  • Sponsorship chair of ICDM 2021
  • Publicity chair of ADC 2023, 2024