Chaveevan Pechsiri and Sumran Painual
Elementary Discourse Unit, medicinal property knowledge extraction, Medicinal Property Boundary
The aim of this paper is to automatically extract the herbal medicinal property knowledge from the web technical documents as knowledge sources for automatically constructing knowledge base which benefits to health-care center for providing the interesting information. The extracted medicinal property knowledge is based on multiple simple sentence or EDUs (Elementary Discourse Units). There are three problems of ex-tracting the medicinal property knowledge: the herbal object identification, the medicinal property identification, and the medicinal property boundary de-termination. We propose using machine learning technique as SVM and NLP (Natural Language Processing) including Centering Theory to solve all those three problems. The result of extraction shows successfully the precision and re-call of 85% and 70%, respectively, with 90% correctness of the boundary determination.
Important Links:
Go Back
