TY - GEN
T1 - Neural Concept Map Generation for Effective Document Classification with Interpretable Structured Summarization
AU - Yang, Carl
AU - Zhang, Jieyu
AU - Wang, Haonan
AU - Li, Bangzheng
AU - Han, Jiawei
N1 - Publisher Copyright:
© 2020 ACM.
PY - 2020/7/25
Y1 - 2020/7/25
N2 - Concept maps provide concise structured representations for documents regarding their important concepts and interaction links, which have been widely used for document summarization and downstream tasks. However, the construction of concept maps often relies heavily on heuristic design and auxiliary tools. Recent popular neural network models, on the other hand, are shown effective in tasks across various domains, but are short in interpretability and prone to overfitting. In this work, we bridge the gap between concept map construction and neural network models, by designing doc2graph, a novel weakly-supervised text-to-graph neural network, which generates concept maps in the middle and is trained towards document-level tasks like document classification. In our experiments, doc2graph outperforms both its traditional baselines and neural counterparts by significant margins in document classification, while producing high-quality interpretable concept maps as document structured summarization.
AB - Concept maps provide concise structured representations for documents regarding their important concepts and interaction links, which have been widely used for document summarization and downstream tasks. However, the construction of concept maps often relies heavily on heuristic design and auxiliary tools. Recent popular neural network models, on the other hand, are shown effective in tasks across various domains, but are short in interpretability and prone to overfitting. In this work, we bridge the gap between concept map construction and neural network models, by designing doc2graph, a novel weakly-supervised text-to-graph neural network, which generates concept maps in the middle and is trained towards document-level tasks like document classification. In our experiments, doc2graph outperforms both its traditional baselines and neural counterparts by significant margins in document classification, while producing high-quality interpretable concept maps as document structured summarization.
KW - document classification
KW - document representation learning
KW - document summarization
KW - graph generation
KW - graph mining
UR - http://www.scopus.com/inward/record.url?scp=85090125535&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85090125535&partnerID=8YFLogxK
U2 - 10.1145/3397271.3401312
DO - 10.1145/3397271.3401312
M3 - Conference contribution
AN - SCOPUS:85090125535
T3 - SIGIR 2020 - Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval
SP - 1629
EP - 1632
BT - SIGIR 2020 - Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval
PB - Association for Computing Machinery
T2 - 43rd Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2020
Y2 - 25 July 2020 through 30 July 2020
ER -