Title: Entity Linking with graph functional dependencies
Entity resolution (ER) is the problem of accurately identifying multiple, differing, and possibly contradicting representations of unique real-world entities in data. It is a challenging, paramount, and fundamental task in data cleansing and
data integration. In this work, we propose an effective approach to ER using graph differential dependencies (GDDs). GDDs are an extension of the recent graph entity dependencies (which are formal constraints for graph data) to enable approximate matching of values. We propose GDDs, investigate a special discovery for ER by designing an algorithm for generating a non-redundant set of GDDs in labelled data.
Furthermore, we develop an ER technique, Certus, that employs the learnt GDDs for improving the accuracy of ER results.
Dr Jixue Liu's research and teaching areas are mainly in database and data mining areas. His research topics are in XML, dependency theory and discovery, privacy, text and sequence data analytics, data integration and entity resolution. He teaches database course, XML related courses, and data mining courses. He taught programming and Web development courses.