Autor: |
Fabio Garzia, Francesco Borghini, Alberto Bruni, Mara Lombardi, Ludovica Minò, Soodamani Ramalingam, Giorgia Tricarico |
Jazyk: |
angličtina |
Rok vydání: |
2022 |
Předmět: |
|
Zdroj: |
Sensors, Vol 22, Iss 21, p 8138 (2022) |
Druh dokumentu: |
article |
ISSN: |
1424-8220 |
DOI: |
10.3390/s22218138 |
Popis: |
This paper proposes a methodology for sentiment analysis with emphasis on the emotional aspects of people visiting the Herculaneum Archaeological Park in Italy during the period of the COVID-19 pandemic. The methodology provides a valuable means of continuous feedback on perceived risk of the site. A semantic analysis on Twitter text messages provided input to the risk management team with which they could respond immediately mitigating any apparent risk and reducing the perceived risk. A two-stage approach was adopted to prune a massively large dataset from Twitter. In the first phase, a social network analysis and visualisation tool NodeXL was used to determine the most recurrent words, which was achieved using polarity. This resulted in a suitable subset. In the second phase, the subset was subjected to sentiment and emotion mapping by survey participants. This led to a hybrid approach of using automation for pruning datasets from social media and using a human approach to sentiment and emotion analysis. Whilst suffering from COVID-19, equally, people suffered due to loneliness from isolation dictated by the World Health Organisation. The work revealed that despite such conditions, people’s sentiments demonstrated a positive effect from the online discussions on the Herculaneum site. |
Databáze: |
Directory of Open Access Journals |
Externí odkaz: |
|
Nepřihlášeným uživatelům se plný text nezobrazuje |
K zobrazení výsledku je třeba se přihlásit.
|