Sie befinden Sich nicht im Netzwerk der Universität Paderborn. Der Zugriff auf elektronische Ressourcen ist gegebenenfalls nur via VPN oder Shibboleth (DFN-AAI) möglich. mehr Informationen...
In this paper, a rule-based information extraction method of descriptive paragraphs of scenic spots is proposed. While paying attention to the location information of scenic spots, it extracts brief and general descriptive paragraphs of scenic spots which are taken as descriptive texts of scenic spots, which is of academic reference value for textual information extraction beyond the current entity triple. The experiments are conducted respectively from perspective of the location relation, the affiliation, and the relation of creation of time. When only considering the location relation and rules of the affiliation, the accuracy rate is 90.85%, the recall rate is 85.43%, and the F value is 88.06%. So the validity and applicability of this method are proved.