A SURVEY ON KNOWLEDGE BASE REFINEMENT

PhD Qualifying Examination


Title: "A SURVEY ON KNOWLEDGE BASE REFINEMENT"

by

Mr. Hao XIN


Abstract:

Knowledge base is an organized repository of knowledge in computer 
processable form. In the recent years, a lot of knowledge bases(KBs) have 
been created, notable examples including Freebase, DBpedia and Yago. Those 
KBs are often either constructed from semi-structured knowledge, such as 
Wikipedia, or harvested from the web using a combination of statistical 
and linguistic methods. The result are large-scale knowledge bases that 
try to make a good trade-off between completeness and correctness. With 
the increasing popularity and construction of large-scale KBs, various 
kinds of applications have already tried to take advantage of the KBs, 
e.g., semantic search, natural language translation, deep QA systems. 
However as a model of the real world, formalized knowledge base cannot 
reasonably reach full coverage or fully correct. In order to solve this 
problem and further increase the utility of such knowledge bases, a lot of 
refinement methods have been proposed, which try to infer and add missing 
knowledge to the knowledge base, or identify the wrong information. In 
this article, we provide a survey of such knowledge graph refinement 
approaches.


Date:			Tuesday, 26 June 2018

Time:                  	3:00pm - 5:00pm

Venue:                  Room 5560
                         Lifts 27/28

Committee Members:	Prof. Lei Chen (Supervisor)
 			Dr. Ke Yi (Chairperson)
 			Prof. Dimitris Papadias
 			Dr. Yangqiu Song


**** ALL are Welcome ****