TY - GEN
T1 - Explaining missing query results in natural language
AU - Deutch, Daniel
AU - Frost, Nave
AU - Gilad, Amir
AU - Haimovich, Tomer
N1 - Publisher Copyright:
© 2020 Copyright held by the owner/author(s). Published in Proceedings of the 23rd International Conference on Extending Database Technology (EDBT), March 30-April 2, 2020, ISBN 978-3-89318-083-7 on OpenProceedings.org. Distribution of this paper is permitted under the terms of the Creative Commons license CC-by-nc-nd 4.0.
PY - 2020/1/1
Y1 - 2020/1/1
N2 - We propose in this paper a novel approach for explaining query non-answers in Natural Language within the context of Natural Language Interfaces to Databases (NLIDBs). Such interfaces allow non-expert users to pose queries over an underlying database; our goal is to further allow users to ask why some results that they have expected to see, are missing from the output. In a nutshell, our approach is to “marry" NLIDBs with an existing model for explaining missing query results by pinpointing the last query operator that is “responsible" for the missing result. We observe that one can often trace the parts of the original NL question that correspond to these operators. This paves the way for intuitive explanations of the non-answers, that are based on highlighting the relevant parts of the question. Our architecture is generic and is not coupled with a specific NLIDB, and our solution yields clear explanations in interactive speed.
AB - We propose in this paper a novel approach for explaining query non-answers in Natural Language within the context of Natural Language Interfaces to Databases (NLIDBs). Such interfaces allow non-expert users to pose queries over an underlying database; our goal is to further allow users to ask why some results that they have expected to see, are missing from the output. In a nutshell, our approach is to “marry" NLIDBs with an existing model for explaining missing query results by pinpointing the last query operator that is “responsible" for the missing result. We observe that one can often trace the parts of the original NL question that correspond to these operators. This paves the way for intuitive explanations of the non-answers, that are based on highlighting the relevant parts of the question. Our architecture is generic and is not coupled with a specific NLIDB, and our solution yields clear explanations in interactive speed.
UR - http://www.scopus.com/inward/record.url?scp=85084189200&partnerID=8YFLogxK
U2 - 10.5441/002/edbt.2020.49
DO - 10.5441/002/edbt.2020.49
M3 - Conference contribution
AN - SCOPUS:85084189200
T3 - Advances in Database Technology - EDBT
SP - 427
EP - 430
BT - Advances in Database Technology - EDBT 2020
A2 - Bonifati, Angela
A2 - Zhou, Yongluan
A2 - Vaz Salles, Marcos Antonio
A2 - Bohm, Alexander
A2 - Olteanu, Dan
A2 - Fletcher, George
A2 - Khan, Arijit
A2 - Yang, Bin
PB - OpenProceedings.org
T2 - 23rd International Conference on Extending Database Technology, EDBT 2020
Y2 - 30 March 2020 through 2 April 2020
ER -