About Me
👨🎓 I am a Senior Data Scientist at Walmart Global Tech.
- 🏢 Department: Marketing and Demand Science
- 🔥 Interests: Generative AI, LLM, Multi-modal learning, Recommender system, NLP
- 🛠️ Skills: Python, C/C++, PyTorch, Tensorflow, Scikit-learn, Pandas
Research Interests
Natural Language Processing (NLP)
- Large language models (LLM)
- Multi-modal learning
- Self-supervised learning
Healthcare
- Temporal health event prediction
- Structured health data generation
- Clinical note analysis
Graph Neural Networks
- Dynamic graph learning
- Graph structure in health data
- Community discovery
Applied Machine Learning
- Healthcare
- Big data
- Time-series
Education
Work Experience
Project Experience
A news crawler for BBC News, Reuters and New York Times
Dec 2017 - Now
Fudan University & Stevens Institute of Technology
Star- https://github.com/LuChang-CS/news-crawler
- Programming language: Python 3
- Automatically extracting news content and meta data
Semi-Structured ICD Coding (NeurIPS 2023)
Aug 2022 - May 2023
Stevens Institute of Technology
Star- Programming language & framework: Python, PyTorch, Scikit-learn
- Propose to automatically segment clinical nots into sections
- Tree-based contrastive learning for multi-label text classification
Generating Multi-label Time-series EHR Data (IEEE TKDE)
Sep 2021 - Apr 2022
Stevens Institute of Technology
Star- Programming language & framework: Python, PyTorch, Scikit-learn
- Propose a novel multi-label time-series generative adversarial network (GAN) to generate electronic health records (EHR)
- Effectively generate uncommon diseases
Temporal Health Event Prediction
Collaborative Learning for Patient and Disease
StarStevens Institute of Technology
- Programming language & framework: Python, Tensorflow, Scikit-learn
- Propose to collaboratively learn representations of patients and diseases with clinical notes using graph neural networks
- Improve the Recall@10 by ~3%
Jan 2020 - Dec 2020
Publications
-
Towards Semi-Structured Automatic ICD Coding via Tree-based Contrastive LearningNeurIPS 2023. [C] (Acceptance Rate: 26.1%)
-
Multi-Label Clinical Time-Series Generation via Conditional GANIEEE Transactions on Knowledge and Data Engineering. 2023. [J] (IF: 8.9)
-
Context-aware Health Event Prediction via Transition Functions on Dynamic Disease GraphsAAAI 2022. [C] (Acceptance Rate: 15.0%)
-
Self-Supervised Graph Learning with Hyperbolic Embedding for Temporal Health Event PredictionIEEE Transactions on Cybernetics. 2021. [J] (IF: 19.118)
-
Collaborative Graph Learning with Auxiliary Text for Temporal Event Prediction in HealthcareIJCAI 2021. [C] (Acceptance Rate: 13.9%)
-
Efficiently Answering Top-k Frequent Term Queries in Temporal-Categorical RangeInformation Sciences. 2021. [J] (IF: 8.233)
-
A System for Exploratory Analysis in CloudIEEE CSCloud/EdgeCom 2018. [C] (Best Student Paper)
Professional Services
Reviewer for journals
- IEEE Transactions on Network Science and Engineering (IEEE TNSE)
- IEEE Transactions on Big Data (IEEE TBD)
- IEEE Transactions on Cybernetics (IEEE TCYB)
- Information Sciences
- Neural Processing Letters
- Computer
Reviewer for conferences
- The 26th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2022)
- Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI 2023)
- 1st AAAI Workshop on Uncertainty Reasoning and Quantification in Decision Making (UDM-AAAI23)
- Thirty-Seventh AAAI Conference on Artificial Intelligence (AAAI 2024)