Name (abbrev)

Name (full)

Category

Last update

Geoquery

Language

b D, Y

Application domain

Further specifications

Geoquery

Type

Format

Complexity

ILP

Prolog

250 facts

WWW / FTP

ftp://ftp.cs.utexas.edu/pub/mooney/nl-ilp-data

Contact person(s)

Related group(s)

Optional contact address

References

John M. Zelle and Raymond J. Mooney
"Learning to Parse Database Queries Using Inductive Logic Programming"
Submitted to the Fourteenth National Conference
on Artificial Intelligence (AAAI-96)

Annotations

These are data for natural-language learning experiments (Geoquery) in a form suitable for Inductive Logic Programming (ILP) systems. This is a set of data relating English queries about a simple U.S. geography database to executable Prolog queries in the following form:

parse([how,many,people,live,in,hawaii,?],
answer(B,(population(A,B),const(A,stateid(hawaii))))).
ILP is applied to induce rules defining an appropriate natural-language database front-end [Zelle 96]. Details are in ftp://ftp.cs.utexas.edu/pub/mooney/papers/chill-dissertation-95.ps.
This data set is also related to the task described in Natural Language Parsing application domain (M & K data). A summary could be found in .../papers/chill-bkchapter-95.ps

Comments