Development of extensible open information extraction
Elvan Owen, Dwi Hendratmo Widyantoro · 2017
Open information extraction is used to extract information from open-domain documents such as articles in the Internet. Most extraction techniques have common extraction flow starting from collecting documents, preprocessing documents, extracting informations, postprocessing extracted relations and evaluation of extracted relations. Implementation of a new technique requires the user to build the whole process from scratch. Consequently, implementation becomes a complex and time-consuming process. Therefore, a prototype acting as a framework to abstract and split the whole extraction processes into multiple separate processes implemented as components or plugins will allow users to reuse already existing extraction subprocess.