Dr. Qun Chen
Northwestern Polytechnical University
Title: A Case for Risk Analysis: Enabling Quality Control for Entity Resolution.
Abstract: Even though many machine algorithms have been proposed for entity resolution, it remains very challenging to find a solution with quality guarantees. In this talk, I will first present a general HUman and Machine Cooperative (HUMO) framework for entity resolution (ER). By dividing an ER workload between machine and human, HUMO enables a mechanism for quality control that can flexibly enforce both precision and recall levels. Then, I will present an improvement of HUMO, a risk-aware HUman-Machine Cooperation framework for ER, denoted by r-HUMO. r-HUMO is the first solution that optimizes the process of human workload selection from a risk perspective. Finally, I will summarize our work on risk analysis for entity resolution and try to build a case for its value as a future research direction. I will demonstrate how our work on entity resolution can be generalized to other challenging classification tasks.
Short Biography: Qun Chen received his bachelor degree from Tsinghua University, China, in 1998, and his Ph.D degree in computer science from National University of Singapore in 2003. His current research interests mainly focus on the general enabling frameworks and technologies for challenging tasks (e.g. data cleaning and sentiment analysis), which usually require the fusion of a wide variety of techniques, including data analytics, artificial intelligence and human effort. He has been responsible for multiple national projects in related areas, including the 973 sub-project of Ministry of Science and Technology of China, and the key project of NSF of China. He has published more than 60 papers in renowned conferences and journals, including the top-tier SIGMOD, VLDB, ICDE, WWW and TKDE.