Pei Li

I am currently a post-doc in Database Technology Group at University of Zurich. I received my Ph.D education in Computer Science at University of Milan - Bicocca from 2009 to 2012. During my Ph.D study, I visited AT&T Labs – Research in NJ, USA during 2010 and 2012. Previously, I obtained a M.S. in Computer Science at Beijing University of Posts and Telecommunications and a B.S. in Electronic Engineering at the same university. My research interests are in data quality, data cleaning and data integration.

Please find my CV

here. (PDF, 88 KB)

Databases, with an emphasis on data quality, data cleaning and data integration, particularly in record linkage, also interested in large-scale data analytics.

Chronos - Linking records with value diversity

In many data sets records that refer to the same real-world entity can have very diverse values, because of erroneous values, value evolution over time, or "local" properties for members in the same group. We study how to link records with tolerance to fairly high diversity of values.

Talk (PPTX, 1 MB)

Demo

Paper (PDF, 547 KB)

• Database Management and Performance Tuning

Book Chapter

• Pei Li and Andrea Maurino: Record Linkage in Complex Contexts. In “Handbook on Research and Practice in Data Quality”. To be published by Springer.

Journal

• Pei Li, Xin Luna Dong, Andrea Maurino and Divesh Srivastava: Linking temporal records. Frontiers of Computer Science 6(3): 293-312 (2012). (invited)

Papers in Refereed Conferences

• Pei Li, Xin Luna Dong, Songtao Guo, Andrea Maurino and Divesh Srivastava: Robust Group Linkage. In Proceedings of the 24th International Conference on World Wide Web (WWW), 2015, to appear. (acceptance rate 14.1%)
  Paper (PDF, 818 KB)
  Report (PDF, 2 MB)
• Pei Li, Xin Luna Dong, Andrea Maurino and Divesh Srivastava: Linking Temporal Records. In Int. Conf. on Very Large Data Bases (VLDB), pages 956–967, 2011. (acceptance rate 18.1%)

Conference Demo

• Pei Li, Haidong Wang, Christina Tziviskou, Xin Luna Dong, Xiaoguang Liu, Andrea Maurino and Divesh Srivastava: Chronos: Facilitating History Discovery by Linking Temporal Records. In Int. Conf. on Very Large Data Bases (VLDB), 2012.