Construction of Symptom-cause Knowledge Graph based on Named Entity Recognition and Relation Extraction

Pengcheng Zhou, Honghai Feng, Jun Li, Yaju Wei, Ruihui Hou, Yanyan Chen

2022

Abstract

The symptom-cause knowledge graph can provide great helps for doctors and machines to diagnose diseases. However, the existing web texts that include the causes of symptoms are too trivial and complex. This paper uses the entity recognition and relation extraction methods and proposes a algorithm to extract the causes of symptoms by semantic rules. According to the relationships of semantic elements in sentences a knowledge graph of symptom-cause is built. The process is as follows: Firstly, a large number of texts that include cause-symptom are obtained. Secondly, the high-frequency sentence patterns of the symptom-cause relationships are summarized. Thirdly, an algorithm for extracting the cause information of the corresponding symptoms according to the sentence rules are designed. From the obtained 2856331 web pages, 108693 of the causes corresponding to 40236 symptoms are obtained.

Paper Citation

in Harvard Style

Zhou P., Feng H., Li J., Wei Y., Hou R. and Chen Y. (2022). Construction of Symptom-cause Knowledge Graph based on Named Entity Recognition and Relation Extraction. In Proceedings of the 1st International Conference on Health Big Data and Intelligent Healthcare - Volume 1: ICHIH, ISBN 978-989-758-596-8, pages 17-24. DOI: 10.5220/0011200500003438

in Bibtex Style

@conference{ichih22,
author={Pengcheng Zhou and Honghai Feng and Jun Li and Yaju Wei and Ruihui Hou and Yanyan Chen},
title={Construction of Symptom-cause Knowledge Graph based on Named Entity Recognition and Relation Extraction},
booktitle={Proceedings of the 1st International Conference on Health Big Data and Intelligent Healthcare - Volume 1: ICHIH,},
year={2022},
pages={17-24},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011200500003438},
isbn={978-989-758-596-8},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Health Big Data and Intelligent Healthcare - Volume 1: ICHIH,
TI - Construction of Symptom-cause Knowledge Graph based on Named Entity Recognition and Relation Extraction
SN - 978-989-758-596-8
AU - Zhou P.
AU - Feng H.
AU - Li J.
AU - Wei Y.
AU - Hou R.
AU - Chen Y.
PY - 2022
SP - 17
EP - 24
DO - 10.5220/0011200500003438