Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents
The commonly-used large-scale knowledge bases have been facing challenges in open domain question answering tasks which are caused by the loose knowledge association and weak structural logic of triplet-based knowledge. To find a way out of this dilemma, this work proposes a novel metaknowledge enha...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-12-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/24/8439 |
_version_ | 1797500766785634304 |
---|---|
author | Shukan Liu Ruilin Xu Li Duan Mingjie Li Yiming Liu |
author_facet | Shukan Liu Ruilin Xu Li Duan Mingjie Li Yiming Liu |
author_sort | Shukan Liu |
collection | DOAJ |
description | The commonly-used large-scale knowledge bases have been facing challenges in open domain question answering tasks which are caused by the loose knowledge association and weak structural logic of triplet-based knowledge. To find a way out of this dilemma, this work proposes a novel metaknowledge enhanced approach for open domain question answering. We design an automatic approach to extract metaknowledge and build a metaknowledge network from Wiki documents. For the purpose of representing the directional weighted graph with hierarchical and semantic features, we present an original graph encoder GE4MK to model the metaknowledge network. Then, a metaknowledge enhanced graph reasoning model MEGr-Net is proposed for question answering, which aggregates both relational and neighboring interactions comparing with R-GCN and GAT. Experiments have proved the improvement of metaknowledge over main-stream triplet-based knowledge. We have found that the graph reasoning models and pre-trained language models also have influences on the metaknowledge enhanced question answering approaches. |
first_indexed | 2024-03-10T03:08:34Z |
format | Article |
id | doaj.art-4e8efa7bf3474b48827024463cce94fe |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T03:08:34Z |
publishDate | 2021-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-4e8efa7bf3474b48827024463cce94fe2023-11-23T10:31:29ZengMDPI AGSensors1424-82202021-12-012124843910.3390/s21248439Metaknowledge Enhanced Open Domain Question Answering with Wiki DocumentsShukan Liu0Ruilin Xu1Li Duan2Mingjie Li3Yiming Liu4School of Computer Science and Engineering, Southeast University, Nanjing 211189, ChinaSchool of Electronic Engineering, PLA Naval University of Engineering, Wuhan 430033, ChinaSchool of Electronic Engineering, PLA Naval University of Engineering, Wuhan 430033, ChinaShip Comprehensive Test and Training Base, PLA Naval University of Engineering, Wuhan 430033, ChinaSchool of Electronic Engineering, PLA Naval University of Engineering, Wuhan 430033, ChinaThe commonly-used large-scale knowledge bases have been facing challenges in open domain question answering tasks which are caused by the loose knowledge association and weak structural logic of triplet-based knowledge. To find a way out of this dilemma, this work proposes a novel metaknowledge enhanced approach for open domain question answering. We design an automatic approach to extract metaknowledge and build a metaknowledge network from Wiki documents. For the purpose of representing the directional weighted graph with hierarchical and semantic features, we present an original graph encoder GE4MK to model the metaknowledge network. Then, a metaknowledge enhanced graph reasoning model MEGr-Net is proposed for question answering, which aggregates both relational and neighboring interactions comparing with R-GCN and GAT. Experiments have proved the improvement of metaknowledge over main-stream triplet-based knowledge. We have found that the graph reasoning models and pre-trained language models also have influences on the metaknowledge enhanced question answering approaches.https://www.mdpi.com/1424-8220/21/24/8439metaknowledgegraph modelingquestion answeringgraph neural networksknowledge graph |
spellingShingle | Shukan Liu Ruilin Xu Li Duan Mingjie Li Yiming Liu Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents Sensors metaknowledge graph modeling question answering graph neural networks knowledge graph |
title | Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents |
title_full | Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents |
title_fullStr | Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents |
title_full_unstemmed | Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents |
title_short | Metaknowledge Enhanced Open Domain Question Answering with Wiki Documents |
title_sort | metaknowledge enhanced open domain question answering with wiki documents |
topic | metaknowledge graph modeling question answering graph neural networks knowledge graph |
url | https://www.mdpi.com/1424-8220/21/24/8439 |
work_keys_str_mv | AT shukanliu metaknowledgeenhancedopendomainquestionansweringwithwikidocuments AT ruilinxu metaknowledgeenhancedopendomainquestionansweringwithwikidocuments AT liduan metaknowledgeenhancedopendomainquestionansweringwithwikidocuments AT mingjieli metaknowledgeenhancedopendomainquestionansweringwithwikidocuments AT yimingliu metaknowledgeenhancedopendomainquestionansweringwithwikidocuments |