Humans Outperform Machines at the Bilingual Shannon Game

We provide an upper bound for the amount of information a human translator adds to an original text, i.e., how many bits of information we need to store a translation, given the original. We do this by creating a Bilingual Shannon Game that elicits character guesses from human subjects, then develop...

Full description

Bibliographic Details
Main Authors: Marjan Ghazvininejad, Kevin Knight
Format: Article
Language:English
Published: MDPI AG 2016-12-01
Series:Entropy
Subjects:
Online Access:http://www.mdpi.com/1099-4300/19/1/15
id doaj-8d842893a4b74f8597fd1d5bd3d21e71
record_format Article
spelling doaj-8d842893a4b74f8597fd1d5bd3d21e712020-11-24T23:11:59ZengMDPI AGEntropy1099-43002016-12-011911510.3390/e19010015e19010015Humans Outperform Machines at the Bilingual Shannon GameMarjan Ghazvininejad0Kevin Knight1Information Sciences Institute, University of Southern California, 4676 Admiralty Way #1001, Marina Del Rey, CA 90292, USAInformation Sciences Institute, University of Southern California, 4676 Admiralty Way #1001, Marina Del Rey, CA 90292, USAWe provide an upper bound for the amount of information a human translator adds to an original text, i.e., how many bits of information we need to store a translation, given the original. We do this by creating a Bilingual Shannon Game that elicits character guesses from human subjects, then developing models to estimate the entropy of those guess sequences.http://www.mdpi.com/1099-4300/19/1/15compressionmultilingualtranslation
collection DOAJ
language English
format Article
sources DOAJ
author Marjan Ghazvininejad
Kevin Knight
spellingShingle Marjan Ghazvininejad
Kevin Knight
Humans Outperform Machines at the Bilingual Shannon Game
Entropy
compression
multilingual
translation
author_facet Marjan Ghazvininejad
Kevin Knight
author_sort Marjan Ghazvininejad
title Humans Outperform Machines at the Bilingual Shannon Game
title_short Humans Outperform Machines at the Bilingual Shannon Game
title_full Humans Outperform Machines at the Bilingual Shannon Game
title_fullStr Humans Outperform Machines at the Bilingual Shannon Game
title_full_unstemmed Humans Outperform Machines at the Bilingual Shannon Game
title_sort humans outperform machines at the bilingual shannon game
publisher MDPI AG
series Entropy
issn 1099-4300
publishDate 2016-12-01
description We provide an upper bound for the amount of information a human translator adds to an original text, i.e., how many bits of information we need to store a translation, given the original. We do this by creating a Bilingual Shannon Game that elicits character guesses from human subjects, then developing models to estimate the entropy of those guess sequences.
topic compression
multilingual
translation
url http://www.mdpi.com/1099-4300/19/1/15
work_keys_str_mv AT marjanghazvininejad humansoutperformmachinesatthebilingualshannongame
AT kevinknight humansoutperformmachinesatthebilingualshannongame
_version_ 1725602965418409984