In the context of the NewsEye project where many languages are considered, the DAniEL system, created by Gaël Lejeune, was chosen. The system focuses on Epidemic Surveillance over press articles across multiple languages. Rather depending on language-specific grammars and analyzers, the system implements a string-based algorithm that detects repeated maximal substrings in salient zones to extract important information (disease name, location). The decision of choosing prominent text zones is based on the characteristics of a journalistic writing style where crucial information is usually put at the beginning and end of the article.
This is the Python 3 version of the Daniel System (originally from here: https://github.com/rundimeco/daniel)
Neural Methods for Event Extraction https://tel.archives-ouvertes.fr/tel-01943841/document