Traditional Chinese Medicine utilization has rapidly increased worldwide. However, there is limited database provides the information of TCM herbs and diseases. The study aims to identify and evaluate the meaningful associations between TCM herbs and breast cancer by using the association rule mining (ARM) techniques. We employed the ARM techniques for 19.9 million TCM prescriptions by using Taiwan National Health Insurance claim database from 1999 to 2013. 364 TCM herbs-breast cancer associations were derived from those prescriptions and were then filtered by their support of 20. Resulting of 296 associations were evaluated by comparing to a gold-standard that was curated information from Chinese-Wikipedia with the following terms, cancer, tumor, malignant. All 14 TCM herbs-breast cancer associations with their confidence of 1% were valid when compared to gold-standard. For other confidences, the statistical results showed consistently with high precisions. We thus succeed to identify the TCM herbs-breast cancer associations with useful techniques.

Original languageEnglish
Title of host publicationBuilding Continents of Knowledge in Oceans of Data
Subtitle of host publicationThe Future of Co-Created eHealth - Proceedings of MIE 2018
PublisherIOS Press
Number of pages5
ISBN (Electronic)9781614998518
Publication statusPublished - Jan 1 2018
Event40th Medical Informatics in Europe Conference, MIE 2018 - Gothenburg, Sweden
Duration: Apr 24 2018Apr 26 2018

Publication series

NameStudies in Health Technology and Informatics
ISSN (Print)0926-9630
ISSN (Electronic)1879-8365


Other40th Medical Informatics in Europe Conference, MIE 2018


  • Association rule mining
  • Cancers
  • Traditional Chinese Medicine

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics
  • Health Information Management


Dive into the research topics of 'An automated technique to construct a knowledge base of traditional Chinese herbal medicine for cancers: An exploratory study for breast cancer'. Together they form a unique fingerprint.

Cite this