Loading…

Investigating global language networks using Google search queries

•A novel approach to extract and visualize information from Google search engine.•Proposing translation requests as the dataset to generate multiple GLNs.•Evaluating languages position in the network with respect to different criteria.•Generating various statistical hypothesis tests based on represe...

Full description

Saved in:
Bibliographic Details
Published in:Expert systems with applications 2019-05, Vol.121, p.66-77
Main Authors: Esmaeili Aliabadi, Danial, Avşar, Bihter, Yousefnezhad, Reza, Esmaeili Aliabadi, Edris
Format: Article
Language:English
Subjects:
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:•A novel approach to extract and visualize information from Google search engine.•Proposing translation requests as the dataset to generate multiple GLNs.•Evaluating languages position in the network with respect to different criteria.•Generating various statistical hypothesis tests based on represented visuals. Over the centuries, languages have evolved and influenced each other. These changes might have been caused by long wars, economic superiority, religious imposition or scientific domination. These macro-level forces at micro-level deeply involve individuals who want to translate their messages into another language. Therefore, the frequency of translation requests from a language to another might be an acceptable measure to show the necessity of those languages to each other given speakers of those languages are unable to communicate due to linguistic barriers. In this paper, we first collect the average number of translation requests per month from the Google search engine for one hundred languages and then visualize them by using customized presentation software to identify features and patterns. Based on the extracted information, the rank and the position of each language in the global language network have been investigated with respect to four criteria: translation balance, centrality, the number of connections and native speaker population, and interconnections between language families. Our study sheds light on the individual’s consumption behavior of machine translation in different cultures. Our findings might be interesting especially for policymakers who design school curricula.
ISSN:0957-4174
1873-6793
DOI:10.1016/j.eswa.2018.12.016