Prof. Hinrich Schütze


Institut für Maschinelle Sprachverarbeitung, Universität Stuttgart


Thema: The Future of Natural Language Processing: Cognitive Science or Machine Learning?


Termin: 2. Dezember 2005 11:00-12:30 Uhr (im Rahmen des dies academicus), Städt. Kaufhaus Hs 2–07


Betreuung: Prof. Heyer


Finanzierung: Institutsmittel


Abstract:


Language and speech technology have advanced rapidly in the last decades. This is mainly due to the large volumes of text and speech that are now stored electronically as well as the increased computing power available to train statistical models on large data sets. However, the models used are often surprisingly naive from a linguistic point of view. Only very basic insights about language are incorporated. In this talk, we will explore the reasons for this state of affair and look into a crystal ball: Will future models in language processing continue to be simplistic or will they mimick the way humans process language?


Link: http://www.ims.uni-stuttgart.de/~schuetze



 
Zu dieser Seite gibt es keine Dateien. [Zeige Dateien/Upload]
Kein Kommentar. [Zeige Kommentare]