Named Entity recognition and concept mapping for digital press articles in the context of university
In a university context, the Press office plays a role as a connector between the university and the public, as well as an information bridge for its university and students, staff members. In order to do this, the Press office is responsible for publishing digital articles on the university, especially on research and studies. Currently, in TU Chemnitz, the new articles are published and connected with each other by tagging with simple keywords (e.g., name of faculties) or too general keywords (e.g., “Economy”, “Research”). This makes it hard to display all articles related to a particular professorship or person.
The objective of this master thesis is to find an approach or a combination of approaches to automatically identify and extract the mentioned entities (e.g., particular staff members and professorships) in its text content within the article and connect them with other systems. This particularly includes the state of the art regarding named entity recognition and concept mapping. The demonstration of feasibility with an implementation prototype of the concept is part of this thesis as well as a suitable evaluation with exemplary use cases.