Home > Seminars > Chengkai Li - Restoring Trust by Computing: Data-driven Fact-checking and Exceptional Fact Finding

Chengkai Li - Restoring Trust by Computing: Data-driven Fact-checking and Exceptional Fact Finding


4/26/2018 at 3:30PM


4/26/2018 at 4:30PM


126 DeBartolo


College of Engineering close button

Dong Wang

Dong Wang

VIEW FULL PROFILE Email: dwang5@nd.edu
Phone: 574-631-3749
Website: http://www.nd.edu/~dwang5
Office: 214B Cushing


College of Engineering Assistant Professor
Big Data Analytics, Cyber-Physical Systems (CPS), Social Sensing, Smart Cities, Internet of Things (IoT), Network Science
Click for more information about Dong
Add to calendar:
iCal vCal

Our society is struggling with an unprecedented amount of falsehoods. Human fact-checkers cannot keep up with the amount of misinformation and disinformation and the speed at which they spread. This challenge creates an opportunity for automated fact-checking systems. We have been building ClaimBuster, an end-to-end system for data-driven fact-checking. ClaimBuster uses machine learning, natural language processing, and database query techniques to aid in the process of fact-checking. It monitors live discourses and social media to spot factual claims, detect matches between the claims and a curated repository of fact-checks, and deliver those matches instantly to the audience. For various types of new claims not checked before, ClaimBuster automatically translates them into queries against knowledge databases and reports whether they check out. While the development of the full-fledged system is still ongoing, several components of ClaimBuster are integrated and deployed at http://idir.uta.edu/claimbuster/. The project is part of the IDIR Lab's inter-disciplinary research program in computational journalism. Under this program, we have also extensively worked on frameworks, algorithms, and systems for efficient discovery of data-backed facts. Particularly, we developed Maverick, an extensible graph mining framework that discovers exceptional facts about entities in knowledge graphs, and FactWatcher (http://idir.uta.edu/factwatcher/), a suite of fact-finding algorithms that monitor structured event databases. The IDIR Lab is also conducting research on tackling usability challenges in querying, mining, and exploring entity graphs. This talk will first present a high-level overview of these projects, followed by brief discussion of ClaimBuster's claim-spotting method and several ongoing directions. It will then provide more details about FactWatcher and Maverick, which represents a synergy of several research thrusts. Finally, the talk will discuss our plan for research on improving transparency and trust in fact-checking and improving usability of fact-finding systems.

Seminar Speaker:

Chengkai Li

Chengkai Li

The University of Texas at Arlington

Dr. Chengkai Li is an Associate Professor and Associate Chair in the Department of Computer Science and Engineering at the University of Texas at Arlington. He is the Director of the Innovative Database and Information Systems Research Laboratory (IDIR). Dr. Li's research interests span several areas related to big data intelligence and data science, including databases, data mining and applied machine learning, natural language processing, and their applications in computational journalism. His current research focuses on building large-scale human-assisting and human-assisted data and information systems with high usability, high efficiency and applications for social good. Particularly, his ongoing research projects include data-driven fact-checking, exceptional fact finding, fake-news detection, usability challenges in querying and exploring graph data, knowledge databases, and data exploration by ranking (top-k), skyline and preference queries. He has also worked on crowdsourcing, entity queries, database testing, database engine for top-k queries, and Web and XML data management. His publications received several awards at prestigious conferences such as SIGMOD, VLDB, and CIDR. Dr. Li created the inter-disciplinary research program in computational journalism at UTA. He led the development of a fact-checking system ClaimBuster which is well-known in the fact-checking community and has received substantial media coverage. The ClaimBuster API is being used by the Duke Reporters’ Lab to create daily newsletters that recommend the most check-worthy claims to The Washington Post, PolitiFact, and other professional fact-checkers. Dr. Li is also a recipient of the HP Labs Innovation Research Awards in 2011 and 2012. He received his Ph.D. degree in computer science from the University of Illinois at Urbana-Champaign. He graduated from Nanjing University with an M.Eng. degree and a B.S. degree in Computer Science.