Knowledge-graph question answering can fail even when a question is interpreted correctly and its entities and relation types match the graph. If the knowledge graph lacks the fact required to answer the question, the system cannot retrieve the correct answer. This limitation motivates knowledge graph completion and link prediction.

University of Michigan - Ann Arbor

Google

because knowledge graphs are typically incomplete, it is useful to perform knowledge graph completion or link prediction, i.e. predict whether a relationship not in the knowledge graph is likely to be true.

When to use the Knowledge Graphs


The system builds a semantic representation of the query and then use it to query databases of facts

Knowledge-based Question Answering (QA)

How Knowledge Graph Incompleteness Limits Question Answering

+ An RDF triple is a 3-tuple, a predicate with two arguments, expressing
some simple relation or proposition.
+ Can be used to answer text questions like “When was Ada Lovelace
born?” or “Who was born in 1815?”.




Knowledge-Based QA from RDF triple stores (Graph-based QA)

In knowledge-based question answering, semantic parsing maps a natural-language question to a structured program that can be used to retrieve an answer. The resulting logical form may use predicate calculus or a query language such as SQL or SPARQL. Training may be fully supervised with question–logical-form pairs or weakly supervised with question–answer pairs.

Learn Before

Related