"Experiments on generating questions about facts" by Vasile Rus, Zhiqiang Cai et al.
 

Experiments on generating questions about facts

Abstract

This paper presents an approach to the problem of factual Question Generation. Factual questions are questions whose answers are specific facts: who?, what?, where?, when?. We enhanced a simple attribute-value (XML) language and its interpretation engine with context-sensitive primitives and added a linguistic layer deep enough for the overall system to score well on user satisfiability and the 'linguistically well-founded' criteria used to measure up language generation systems. Experiments with open-domain question generation on TREC-like data validate our claims and approach. © Springer-Verlag Berlin Heidelberg 2007.

Publication Title

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

Plum Print visual indicator of research metrics
PlumX Metrics
  • Citations
    • Citation Indexes: 21
  • Usage
    • Abstract Views: 2
  • Captures
    • Readers: 23
see details

Share

COinS