Sentiment Analysis to support book selection: a study applied to the Skoob platform.
Nenhuma Miniatura disponível
Data
2022-01-03
Autores
Pereira, Clayton Martins [UNESP]
Marinho, Ronnie Shida [UNESP]
Segundo, José Eduardo Santarem
Título da Revista
ISSN da Revista
Título de Volume
Editor
Resumo
Objective: This paper aims to apply the sentiment analysis technique to reviews published on the Skoob platform in order to propose a new evaluation parameter to help users in their decision-making about whether or not to read a book. Methods: Exploratory research with a quantitative and qualitative approach, which used, to perform the sentiment analysis, the polarity detection technique, in order to automate the detection of the polarity degree of the opinions contained in the reviews, which can be positive, negative or neutral. A total of 45,114 reviews related to the 20 most read books among Skoob platform users were selected. Results: The obtained results show the potential of applying sentiment analysis to the book reviews as another tool to help the Skoob platform user in his decision making about which book to start reading or which books to put on his list of next reads. Conclusions: Book reviews are important inputs in a social network for readers, since they can influence the reading preferences of its users, as well as present the positive and negative characteristics of a given book. Applying Sentiment Analysis to the opinions contained in such reviews can provide indicators in an automated and fast way, making it possible to gauge users' behavior towards the books they have read, as well as being used as an alternative metric for book evaluation.
Descrição
Palavras-chave
Alternative metrics, Polarity detection, Readers' social networks, Sentiment analysis
Como citar
Encontros Bibli, v. 27.