English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Using lexical language models to detect borrowings in monolingual wordlists

Miller, J. E., Tresoldi, T., Zariquiey, R., Castañón, C. A. B., Morozova, N., & List, J.-M. (2020). Using lexical language models to detect borrowings in monolingual wordlists. PLoS One, 15(12): e0242709. doi:10.17613/m051-e049.

Item is

Files

show Files
hide Files
:
shh2701.pdf (Publisher version), 3MB
Name:
shh2701.pdf
Description:
OA
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-
:
shh2701pre.pdf (Preprint), 3MB
Name:
shh2701pre.pdf
Description:
OA (HC)
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/pdf / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-

Locators

show
hide
Locator:
S1 Table (Supplementary material)
Description:
Detection results by language for seeded borrowings. - (last seen Jan. 2021)
OA-Status:
Locator:
S2 Table (Supplementary material)
Description:
Ten-fold cross validation of detection results by language for WOLD wordlists. - (last seen Jan. 2021)
OA-Status:

Creators

show
hide
 Creators:
Miller, John E., Author
Tresoldi, Tiago1, 2, Author           
Zariquiey, Roberto, Author
Castañón, César A. Beltrán, Author
Morozova, Natalia2, Author           
List, Johann-Mattis1, 2, Author           
Affiliations:
1CALC, Max Planck Institute for the Science of Human History, Max Planck Society, ou_2385703              
2Linguistic and Cultural Evolution, Max Planck Institute for the Science of Human History, Max Planck Society, ou_2074311              

Content

show
hide
Free keywords: Language, Markov models, Semantics, Neural networks, Phonology, Memory recall, Evolutionary linguistics, Recurrent neural networks
 Abstract: Native speakers are often assumed to be efficient in identifying whether a word in their language has been borrowed, even when they do not have direct knowledge of the donor language from which it was taken. To detect borrowings, speakers make use of various strategies, often in combination, relying on clues such as semantics of the words in question, phonology and phonotactics. Computationally, phonology and phonotactics can be modeled with support of Markov n-gram models or -- as a more recent technique -- recurrent neural network models. Based on a substantially revised dataset in which lexical borrowings have been thoroughly annotated for 41 different languages of a large typological diversity, we use these models to conduct a series of experiments to investigate their performance in borrowing detection using only information from monolingual wordlists. Their performance is in many cases unsatisfying, but becomes more promising for strata where there is a significant ratio of borrowings and when most borrowings originate from a dominant donor language. The recurrent neural network performs marginally better overall in both realistic studies and artificial experiments, and holds out the most promise for continued improvement and innovation in lexical borrowing detection. Phonology and phonotactics, as operationalized in our lexical language models, are only a part of the multiple clues speakers use to detect borrowings. While improving our current methods will result in better borrowing detection, what is needed are more integrated approaches that also take into account multilingual and cross-linguistic information for a proper automated borrowing detection.

Details

show
hide
Language(s): eng - English
 Dates: 2020-12-09
 Publication Status: Published online
 Pages: 23
 Publishing info: -
 Table of Contents: Introduction
- Problem and motivation
- State of the art

Materials and methods
- Materials
- Lexical language models
-- Bag of sounds
-- Markov Model
-- Recurrent neutral network
- Decision preocedures
- Assessing detection performance
- Implementation
- Experiments and studies
- Detection of artificially seeded borrowings
- Borrowing detection on real language data
- Factors that influence borrowing detection
- Detecting borrowings from a single donor language
- Comparing entropy distributions

Discussion
- Artificially seeded borrowings
- Borrowing detection on real language data
- Factors influencing borrowing detection
- Detecting borrowings from a single donor language
- Comparing entropy distributions


Conclusion
 Rev. Type: Peer
 Identifiers: DOI: 10.17613/m051-e049
Other: shh2701
 Degree: -

Event

show

Legal Case

show

Project information

show hide
Project name : CALC
Grant ID : 715618
Funding program : Horizon 2020 (H2020)
Funding organization : European Commission (EC)

Source 1

show
hide
Title: PLoS One
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: San Francisco, CA : Public Library of Science
Pages: - Volume / Issue: 15 (12) Sequence Number: e0242709 Start / End Page: - Identifier: ISSN: 1932-6203
CoNE: https://pure.mpg.de/cone/journals/resource/1000000000277850

Source 2

show
hide
Title: Humanities Commons
  Abbreviation : HC
Source Genre: Journal
 Creator(s):
Affiliations:
Publ. Info: New York : Modern Language Association
Pages: - Volume / Issue: - Sequence Number: m051-e049 Start / End Page: - Identifier: URN: https://hcommons.org/