Research on Summary Sentences Extraction Oriented to Chinese Patent

DOI

10.2991/icmeit-19.2019.61How to use a DOI?

Keywords

Chinese patent summarization; learning to rank; semantic weigh; word2vec.

Abstract

In this paper, it describes our system oriented to single document summarization task at Chinese patent. We treat the task as a typical document summarization based on sentence extraction and decide to formulate the task in a supervised learning to rank framework, utilizing both common sentence features including term frequency, sentence position, sentence length for generic document summarization and specially designed semantic weigh feature. Summary sentence are selected according the scores by the LTR model we trained from the patent specification. Evaluation results show that our method is indeed appropriate for this task, outperforming several baseline methods in different aspects.

Copyright

© 2019, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Lei Wang
AU  - Xueqiang Lv
AU  - Xindong You
PY  - 2019/04
DA  - 2019/04
TI  - Research on Summary Sentences Extraction Oriented to Chinese Patent
BT  - Proceedings of the 3rd International Conference on Mechatronics Engineering and Information Technology (ICMEIT 2019)
PB  - Atlantis Press
SP  - 372
EP  - 376
SN  - 2352-538X
UR  - https://doi.org/10.2991/icmeit-19.2019.61
DO  - 10.2991/icmeit-19.2019.61
ID  - Wang2019/04
ER  -