Ph.D. Candidate

Data Mining Research Group
Database and Information Systems Laboratory
Department of Computer Science
University of Illinois at Urbana-Champaign


my CV (updared May, 2016)

Google Scholar | LinkedIn | GitHub

About Me

I'm a final year Ph.D. student in CS@Illinois, a member of Data Mining Research Group leaded by Prof. Jiawei Han, and a Google PhD Fellow. Before that, I received Bachelor degree from Computer Science Department and Chu Kochen Honors College, Zhejiang University in 2012.

I interned at Microsoft Research, Redmond and Microsoft Research, Asia multiple times, and will be visiting Pinterest this coming summer! I was born in Shenzhen, China, a beautiful coastal city situated to Hongkong.

Research Interests

I enjoy discovering principles and developing computational models for knowledge acquisition from text data and structure analysis on linked data. I like to design data-driven methods to discover semantic structures (e.g., entities, scientific concepts, clusters, relationships) from massive text corpora.

What's New

  • July. 2016 - Three papers accepted in EMNLP 2016, CIKM 2016, and NAACL 2016.
  • May 2016 - Our latest embedding tool for Label Noise Reduction in Entity Typing is available at GitHub. The paper has been accepted to KDD 2016.
  • Apr. 2016 - Received C. W. Gear Outstanding Graduate Student Award from CS@Illinois---the highest honor given to one grad. student every year.
  • Mar 2016 - Thrilled to be one of the 39 Google PhD Fellows around the globe, and the sole winner in Structured Data. [CS@ILLINOIS]
  • Mar 2016 - Tutorial on "Automatic Entity Recognition and Typing in Massive Text Data" is accepted in SIGMOD 2016. Joint work with Ahmed El-Kishky, Heng Ji, and Jiawei Han.
  • Dec 2015 - Tutorial on "Automatic Entity Recognition and Typing in Massive Text Corpora" is accepted in WWW 2016. See you in Montreal!
  • August 2015 - We (Team Cornfield) are the runner-up in Prostate Cancer DREAM Challenge. [Press release I] Press release II].
  • August 2015 - ClusType, an automatic entity recognition and typing tool (no human label required!) designed for massive, domain-specifc corpora, is available now. A tutorial and a research paper were presented in KDD'15.

  • Last Modified: 08/25/2016