TY - GEN
T1 - Controlling information aggregation for complex question answering
AU - Kwon, Heeyoung
AU - Trivedi, Harsh
AU - Jansen, Peter
AU - Surdeanu, Mihai
AU - Balasubramanian, Niranjan
N1 - Publisher Copyright:
© Springer International Publishing AG, part of Springer Nature 2018.
PY - 2018
Y1 - 2018
N2 - Complex question answering, the task of answering complex natural language questions that rely on inference, requires the aggregation of information from multiple sources. Automatic aggregation often fails because it combines semantically unrelated facts leading to bad inferences. This paper proposes methods to address this inference drift problem. In particular, the paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Empirical evaluation on an elementary science exam benchmark shows that the proposed methods enables effective aggregation even over larger graphs and demonstrates the complementary value of information aggregation for answering complex questions.
AB - Complex question answering, the task of answering complex natural language questions that rely on inference, requires the aggregation of information from multiple sources. Automatic aggregation often fails because it combines semantically unrelated facts leading to bad inferences. This paper proposes methods to address this inference drift problem. In particular, the paper develops unsupervised and supervised mechanisms to control random walks on Open Information Extraction (OIE) knowledge graphs. Empirical evaluation on an elementary science exam benchmark shows that the proposed methods enables effective aggregation even over larger graphs and demonstrates the complementary value of information aggregation for answering complex questions.
UR - http://www.scopus.com/inward/record.url?scp=85044477982&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044477982&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-76941-7_72
DO - 10.1007/978-3-319-76941-7_72
M3 - Conference contribution
AN - SCOPUS:85044477982
SN - 9783319769400
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 750
EP - 757
BT - Advances in Information Retrieval - 40th European Conference on IR Research, ECIR 2018, Proceedings
A2 - Azzopardi, Leif
A2 - Pasi, Gabriella
A2 - Hanbury, Allan
A2 - Piwowarski, Benjamin
PB - Springer-Verlag
T2 - 40th European Conference on Information Retrieval, ECIR 2018
Y2 - 26 March 2018 through 29 March 2018
ER -