Estimation of Digital Public Science Communication Infrastructure (DPSCI) available in different languages, multilingualism in science communication and quantitative analysis of web performance of AI powered, Neural Machine Translation (NMT)-based ed

Science is central to human development and prosperity. This requires communicating wonders and new ideas in science to general people so as to cultivate scientific temper, particularly among motivated young minds to inspire them to take up career in advancement of science. However, since English is lingua franca of science and about 75% of the humanity do not speak English at all (only 6% are native English speakers), there is a significant language barrier in communicating and accessing science for the non-English and non-native English speakers. Multilingualism has been advocated and concerted efforts have been made to address language barriers. Recent developments in ICT including AI-powered machine translations has enabled motivated readers to access a topic through web search in the chosen language. Importantly, search engines actively push contents to the readers who are otherwise unaware of the topic based on interests and profile. This comes handy in disseminating science to interested audience through smart devices, so long as contents are published using crawlable, unique article - level Uniform Resource Locators (URLs) and are optimised for the search engines for the relevant keywords in that language. This benefit is unavailable to issue-level pdf files and print issues. This study estimates digital public science communication infrastructure (DPSCI) available to the speakers of different languages. For this study, DPSCI means availability of scientific databases & repositories, peer-reviewed scientific journals, institutional press releases, research news from academia, dedicated science news agencies (SNAs) and science magazines/portals/platforms/websites in a language. This study also examines current state of multilingualism in science communication and analyses web search performance of machine translated editions of a multilingual science portal, namely Scientific European using data from its Goggle Search Console (GSC) and Bing Webmaster Tool (BWT) accounts. We demonstrate that many languages spoken by large number of people have grossly inadequate science communication infrastructure. They fall in “very high need” or “high need” of infrastructure development. These “very high need” or “high need” category of languages scoring low on DPSCI system are spoken mainly in Africa, South Asia and some part of South-East Asia and were spoken by about 1.2 billion native speakers. Web search performance of machine translated editions of multilingual online science magazine investigated in this study reveals considerable engagements by the non-English speakers. However, engagements by most of the “very high need” (red) or “high need” (orange) category languages were disproportionately lower than languages with higher DPSCI scores. Incidentally, most of the “very high need” (red) or “high need” (orange) category languages, the machine translated editions of the online science magazine under investigation in this study are the only such scientific platform available to the native speakers. Overall, the analysis indicates that Neural Machine Translation (NMT)-based editions of online science magazine using crawlable, unique article-level URLs and regional domains/subdomains architecture contributes in overcoming language barriers in science and in meeting needs of unserved and underserved languages. Upgrading science communication infrastructure and regularly disseminating articles optimised for search engines and social media in native languages to reach out conveying latest information and new ideas in science to the general audiences, particularly younger is imperative for both science and society. This is significant also in view of the fact that global average rate of social media usage is 58.5% (Lausanne Movement, 2026), these social media users are generally younger (Shahghasemi et al., 2025), the social media usage is associated with poorer mind health outcomes in young adulthood (Thiagarajan et al., 2025), and global prevalence of social media addiction among the university student population is 18.4% (Salari et al., 2025). (Note: This document is a preprint (Version 1.0) prepared for submission to peer-reviewed journals).

Authors

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-17
DOI
https://doi.org/10.5281/zenodo.24300329
Primary Topic
Academic Publishing and Open Access
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

Estimation of Digital Public Science Communication Infrastructure (DPSCI) available in different languages, multilingualism in science communication and quantitative analysis of web performance of AI powered, Neural Machine Translation (NMT)-based ed

Umesh Prasad
1 links
Zenodo (CERN European Organization for Nuclear Research)
Academic Publishing and Open Access
preprint

Estimation of Digital Public Science Communication Infrastructure (DPSCI) available in different languages, multilingualism in science communication and quantitative analysis of web performance of AI powered, Neural Machine Translation (NMT)-based ed

Umesh Prasad
preprint en

Abstract

Science is central to human development and prosperity. This requires communicating wonders and new ideas in science to general people so as to cultivate scientific temper, particularly among motivated young minds to inspire them to take up career in advancement of science. However, since English is lingua franca of science and about 75% of the humanity do not speak English at all (only 6% are native English speakers), there is a significant language barrier in communicating and accessing science for the non-English and non-native English speakers. Multilingualism has been advocated and concerted efforts have been made to address language barriers. Recent developments in ICT including AI-powered machine translations has enabled motivated readers to access a topic through web search in the chosen language. Importantly, search engines actively push contents to the readers who are otherwise unaware of the topic based on interests and profile. This comes handy in disseminating science to interested audience through smart devices, so long as contents are published using crawlable, unique article - level Uniform Resource Locators (URLs) and are optimised for the search engines for the relevant keywords in that language. This benefit is unavailable to issue-level pdf files and print issues. This study estimates digital public science communication infrastructure (DPSCI) available to the speakers of different languages. For this study, DPSCI means availability of scientific databases & repositories, peer-reviewed scientific journals, institutional press releases, research news from academia, dedicated science news agencies (SNAs) and science magazines/portals/platforms/websites in a language. This study also examines current state of multilingualism in science communication and analyses web search performance of machine translated editions of a multilingual science portal, namely Scientific European using data from its Goggle Search Console (GSC) and Bing Webmaster Tool (BWT) accounts. We demonstrate that many languages spoken by large number of people have grossly inadequate science communication infrastructure. They fall in “very high need” or “high need” of infrastructure development. These “very high need” or “high need” category of languages scoring low on DPSCI system are spoken mainly in Africa, South Asia and some part of South-East Asia and were spoken by about 1.2 billion native speakers. Web search performance of machine translated editions of multilingual online science magazine investigated in this study reveals considerable engagements by the non-English speakers. However, engagements by most of the “very high need” (red) or “high need” (orange) category languages were disproportionately lower than languages with higher DPSCI scores. Incidentally, most of the “very high need” (red) or “high need” (orange) category languages, the machine translated editions of the online science magazine under investigation in this study are the only such scientific platform available to the native speakers. Overall, the analysis indicates that Neural Machine Translation (NMT)-based editions of online science magazine using crawlable, unique article-level URLs and regional domains/subdomains architecture contributes in overcoming language barriers in science and in meeting needs of unserved and underserved languages. Upgrading science communication infrastructure and regularly disseminating articles optimised for search engines and social media in native languages to reach out conveying latest information and new ideas in science to the general audiences, particularly younger is imperative for both science and society. This is significant also in view of the fact that global average rate of social media usage is 58.5% (Lausanne Movement, 2026), these social media users are generally younger (Shahghasemi et al., 2025), the social media usage is associated with poorer mind health outcomes in young adulthood (Thiagarajan et al., 2025), and global prevalence of social media addiction among the university student population is 18.4% (Salari et al., 2025). (Note: This document is a preprint (Version 1.0) prepared for submission to peer-reviewed journals).

Zenodo (CERN European Organization for Nuclear Research)
Industry, innovation and infrastructure
Academic Publishing and Open Access
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon