TU Berlin

Fachgebiet Datenbanksysteme und InformationsmanagementScalable Temporal Information Extraction with Data Supply Chain Language Operators

Logo FG DIMA-new  65px


zur Navigation


Bearbeiter: Pawel Tondryk

Betreuer: Jochen Adamek

Angestrebter Abschluss: Master


Within my Master’s thesis, I developed a software that makes it possible to scalable analyze unstructured data (Natural Language Processing). I focused on the extraction of temporal data. The developed system bases on the Stratosphere that is a framework for large scale data analysis. My software component can be used to scalable analyze natural language texts (like news, blogs or forums entries).    

In my Master’s thesis defense I describe the developed program and the technologies used. I also present the evaluation results, especially the results of comparison of HeidelTime- and SUTime-based software components for temporal information extraction. I also present the use case that I developed within my Master’s thesis namely automatic generation of temporal snippets. I present the comparison of two approaches. 



Schnellnavigation zur Seite über Nummerneingabe