Answering Complex Questions by Combining Information from Curated and Extracted Knowledge Bases

Nikita Bhutani, Xinyi Zheng, Kun Qian, Yunyao Li, H. Jagadish


Abstract
Knowledge-based question answering (KB_QA) has long focused on simple questions that can be answered from a single knowledge source, a manually curated or an automatically extracted KB. In this work, we look at answering complex questions which often require combining information from multiple sources. We present a novel KB-QA system, Multique, which can map a complex question to a complex query pattern using a sequence of simple queries each targeted at a specific KB. It finds simple queries using a neural-network based model capable of collective inference over textual relations in extracted KB and ontological relations in curated KB. Experiments show that our proposed system outperforms previous KB-QA systems on benchmark datasets, ComplexWebQuestions and WebQuestionsSP.
Anthology ID:
2020.nli-1.1
Volume:
Proceedings of the First Workshop on Natural Language Interfaces
Month:
July
Year:
2020
Address:
Online
Editors:
Ahmed Hassan Awadallah, Yu Su, Huan Sun, Scott Wen-tau Yih
Venue:
NLI
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1–10
Language:
URL:
https://aclanthology.org/2020.nli-1.1
DOI:
10.18653/v1/2020.nli-1.1
Bibkey:
Cite (ACL):
Nikita Bhutani, Xinyi Zheng, Kun Qian, Yunyao Li, and H. Jagadish. 2020. Answering Complex Questions by Combining Information from Curated and Extracted Knowledge Bases. In Proceedings of the First Workshop on Natural Language Interfaces, pages 1–10, Online. Association for Computational Linguistics.
Cite (Informal):
Answering Complex Questions by Combining Information from Curated and Extracted Knowledge Bases (Bhutani et al., NLI 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.nli-1.1.pdf
Video:
 http://slideslive.com/38929797