Banca de QUALIFICAÇÃO: CAROLINA NEVES SOUZA

Uma banca de QUALIFICAÇÃO de DOUTORADO foi cadastrada pelo programa.
DISCENTE : CAROLINA NEVES SOUZA
DATA : 24/07/2023
HORA: 14:00
LOCAL: online
TÍTULO:

 

Utilizando as mídias sociais para entender o sentimento público em relação aos Parques Nacionais Brasileiros

PALAVRAS-CHAVES:

Culturomics; Protected areas; Sentiment analysis, Wildfires; Twitter.

PÁGINAS: 1
RESUMO:

Protected areas (PAs) play a vital role in the conservation of natural and cultural heritage while
supporting local livelihoods. However, PAs can also be seen as barriers to development due to
competition with other potential land-uses. This is the case in Brazil, where limited resources and poor
effectiveness lead to negative sentiments and are leveraged as criticism towards PAs. There is
therefore a need to better comprehend public perceptions of Brazilian PAs and identify the key factors
contributing to negative perceptions. Here, we use data from online discussions about Brazilian
National Parks on Twitter and sentiment analysis to explore this question. The sentiment classification
task involved employing the BERTimbau Base model, a pre-trained model for Brazilian Portuguese
language, to classify over 100,000 social media posts collected over a twelve-year period (2011-2022)
from Twitter. To ensure model accuracy, an additional dataset comprising user opinions about online
products was utilised to train the model. Moreover, we conducted a topic modelling analysis using the
BERTopic model to discern the predominant subjects discussed in relation to the national parks. We
identified 18,388 (17,30%) posts as depicting negative sentiment towards National Parks. The results
unveiled peaks of negative sentiment associated with wildfires during the period of 2011 to 2017, and
a subsequent amplification of negative sentiment since 2019 in association with concerning
government decisions impacting conservation efforts. The topic analysis revealed six prominent
themes reflecting a diverse range of concerns related to the parks, surpassing the influence of isolated
events. Furthermore, examining specific topics on a per-park basis proved beneficial in identifying
distinct issues and conflicts in the six most frequently visited parks in 2021, thereby facilitating

targeted conservation actions. Using social media data to better understand public perceptions of
national parks can strengthen their management and governance by reinforcing their conservation
initiatives and enhancing visitor experiences. Our findings underscore the value of sentiment analysis
in identifying gaps and driving improvements in the management of protected areas.

MEMBROS DA BANCA:
Interno(a) - 3882870 - ANA CLAUDIA MENDES MALHADO
Interno(a) - 2150648 - RAFAEL RICARDO VASCONCELOS DA SILVA
Externo(a) à Instituição - ALEXANDRE SCHIAVETTI - UESC-BA
Externo(a) à Instituição - CHIARA BRAGAGNOLO