Please use this identifier to cite or link to this item: http://hdl.handle.net/10071/20097
Author(s): Fonseca, A. F.
Bandyopadhyay, S.
Louçã, J.
Manjaly, J.
Date: 2019
Title: Caste in the news – a computational analysis of Indian newspapers
Volume: 5
Number: 4
Pages: 1 - 7
ISSN: 2056-3051
DOI (Digital Object Identifier): 10.1177/2056305119896057
Keywords: Content analysis
Computational social science
Natural language processing
Caste
News media
Abstract: Conflicts involving caste issues, mainly concerning the lowest caste rights, pervade modern Indian society. Caste affiliation, being rigorously enforced by the society, is an official contemporary reality. Although caste identity is a major social discrimination, it also serves as a necessary condition for affirmative action like reservation policy. In this article, we perform an original and rigorous analysis of the discourse involving the theme “caste” in India newspapers. To this purpose, we have implemented a computational analysis over a big dataset of the 2016 and 2017 editions of three major Indian newspapers to determine the most salient themes associated with “caste” in the news. We have used an original mix of state-of-the-art algorithms, including those based on statistical distributions and two-layer neural networks, to detect the relevant topics in the news and characterize their linguistic context. We concluded that there is an excessive association between lower castes, victimization, and social unrest in the news that does not adequately cover the reports on other aspects of their life and personal identity, thus reinforcing conflict, while attenuating the vocality and agency of a large section of the population. From our conclusion, we propose a positive discrimination policy in the newsroom.
Peerreviewed: yes
Access type: Open Access
Appears in Collections:ISTAR-RI - Artigos em revistas científicas internacionais com arbitragem científica

Files in This Item:
File Description SizeFormat 
2056305119896057.pdfVersão Editora404,64 kBAdobe PDFView/Open


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpaceOrkut
Formato BibTex mendeley Endnote Logotipo do DeGóis Logotipo do Orcid 

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.