Open information extraction from the web
IJCAI'07 Proceedings of the 20th international joint conference on Artifical intelligence
Ontology-based information extraction: An introduction and a survey of current approaches
Journal of Information Science
| Hi-index | 0.00 |
Information extraction using Natural Language Processing (NLP) tools focuses on extracting explicitly stated information from textual material. This includes Named Entity Recognition (NER), which produces entities and some of the relationships that may exist among them. Intelligent analysis requires examining the entities in the context of the entire document. While some of the relationships among the recognized entities may be preserved during extraction, the overall context of a document may not be preserved. In order to perform intelligent analysis on the extracted information, we provide an ontology, which describes the domain of the extracted information, in addition to rules that govern the classification and interpretation of added elements. The ontology is at the core of an interactive system that assists analysts with the collection, extraction, organization, analysis and retrieval of information, with the topic of "terrorism financing" as a case study. User interaction provides valuable assistance in assigning meaning to extracted information. The system is designed as a set of tools to provide the user with the flexibility and power to ensure accurate inference. This case study demonstrates the information extraction features as well as the inference power that is supported by the ontology.