Publication Details

Summary report for project Exploiting Language Information for Situational Awareness (ELISA) For year 2017

GLEMBEK Ondřej. Summary report for project Exploiting Language Information for Situational Awareness (ELISA) For year 2017. Brno: University of Southern California, 2017.
Czech title
Souhrnná zpráva za rok 2017 pro projekt ELISA v programu DARPA-LORELEI
Type
summary report
Language
english
Authors
Keywords

automatic speech recognition, topic detection, incidents, situational awareness

Abstract

The report summarizes the results of project "Exploiting Language Information for Situational Awareness (ELISA)" performed for the University of Southern California in L.A. in 2016.

Annotation

Cross lingual topic id in latent space: Initial experiments were conducted on a multi-lingual (EN, FR, DE, IT) corpus of political texts comprising of 7 topics and approximately 4400 samples per language. Classification using uncertainties in the learned representations: We proposed a model that learns the document / sentence embeddings along with the uncertainties.

Published
2017
Pages
1-2
Publisher
University of Southern California
Place
Brno, CZ
Back to top