Bachelor Thesis BCLR-2018-111

BibliographyTorhoff, Juri: Historical Event Retrieval.
University of Stuttgart, Faculty of Computer Science, Electrical Engineering, and Information Technology, Bachelor Thesis No. 111 (2018).
58 pages, english.
Abstract

Th?e task of retrieving relevant past newspaper articles given a name of a historical event is challenging. ?e contemporary names for historical events are usually se?led ”a?er the fact” — i.e. a?er some times passes since the event. ?us, ?nding past newspaper issues given the name of the event is problematic, as they simply do not contain the established, contemporary name known to us. ?e objective of this thesis is to devise an approach that would return relevant newspaper issues given the name of an event. For this purpose, two query expansion methods utilizing the contents of Wikipedia were developed and evaluated on a corpus of 19th-century newspapers. ?e approach shows promising preliminary results slightly outperforming naive approaches.

Department(s)University of Stuttgart, Institute for Natural Language Processing
Superviser(s)Padó, Prof. Sebastian, Klinger, Dr. Roman
Entry dateJune 19, 2019
   Publ. Computer Science