Computer Science – Learning
Scientific paper
Jan 2010
adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2010aas...21543815t&link_type=abstract
American Astronomical Society, AAS Meeting #215, #438.15; Bulletin of the American Astronomical Society, Vol. 42, p.395
Computer Science
Learning
Scientific paper
Until now, there has been no available natural language interfaces (NLI's) for querying a database of pulsars (rotating neutron stars emitting radiation at regular intervals). Currently, pulsar records are retrieved through an HTML form accessible via the Australia Telescope National Facility (ATNF) website where one needs to be familiar with pulsar attributes used by the interface (e.g. BLC). Using a NLI relinquishes the need for learning form-specific formalism and allows execution of more powerful queries than those supported by the HTML form. Furthermore, on database access that requires comparison of attributes for all the pulsar records (e.g. what is the fastest pulsar?), using a NLI for retrieving answers to such complex questions is definitely much more efficient and less error-prone. This poster presents the first NLI ever created for the ATNF pulsar database (ATNF-Query) to facilitate database access using complex queries. ATNF-Query is built using a machine learning approach that induces a semantic parser from a question corpus; the innovative application is intended to provide pulsar researchers or laymen with an intelligent language understanding database system for friendly information access.
Dartez Louis
Jenet Frederick
Rangel S.
Tang Rupert
No associations
LandOfFree
Building a Natural Language Interface for the ATNF Pulsar Database for Speeding up Execution of Complex Queries does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.
If you have personal experience with Building a Natural Language Interface for the ATNF Pulsar Database for Speeding up Execution of Complex Queries, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Building a Natural Language Interface for the ATNF Pulsar Database for Speeding up Execution of Complex Queries will most certainly appreciate the feedback.
Profile ID: LFWR-SCP-O-968806