• Your current position:
  • News/Events >
  • Events >
  • Lecture: "Perspectives on Building a Multilingual Parallel...

Print this Page. Send this Page.

Lecture: "Perspectives on Building a Multilingual Parallel Corpus of Ancient Buddhist Scriptures Based on Machine Learning"

Jun 28, 2018 04:00 pm to 06:00 pm

Lecture by Oliver Hellwig (Düsseldorf University, SFB 991 ohellwig@phil-fak.uni-duesseldorf.de) and Sebastian Nehrdich (University of Hamburg, Numata Center for Buddhist Studies sebastian.nehrdich@studium.uni-hamburg.de)

Due to great advances in the field of machine leaarning in the recent years,
new opportunities are showing up for the task of building a multilingual parallel
corpus of ancient Buddhist scriptures. The talk will consists of two parts: First,
Oliver Hellwig will talk about a new machine learning based approach to tackle
two of the long-standing problems of Sanskrit digital processing: the splitting
of Sandhi and the seperation of compounds. In the second part, based on the
results presented by Oliver Hellwig, Sebastian Nehrdich will talk about how
multilingual word embeddings can be used in order to align digital Sanskrit
texts with their Tibetan counterparts. Since this method is working on a high
level of abstraction, it shows a great potencial to be applied to other languagepairs
such as Sanskrit and Buddhist Chinese and Tibetan and Buddhist Chinese
as well.

Search

RSS Feeds

RSS FeedEvent Feed