Análise de sentimentos da performance do técnico do Coritiba Futebol Clube no Campeonato Paranaense de Futebol
Resumo
Resumo : O Youtube é uma das redes sociais mais utilizadas pelos brasileiros e o potencial de divulgação de conteúdo através de vídeos fez com que os clubes de futebol brasileiros dessem uma maior atenção a plataforma, a fim de estabelecerem uma melhor relação com seus torcedores. O trabalho tem como objetivo geral identificar e analisar os sentimentos dos comentários de vídeos da plataforma do Youtube, no âmbito da performance do técnico de futebol do Coritiba Futebol Clube, em sua trajetória de começo de temporada no Campeonato Paranaense de futebol de 2023. Como metodologia, foi definido a linguagem Python e algumas bibliotecas de apoio como a ferramenta a ser utilizada para os processos de coleta, pré-processamento e análise dos dados extraídos da rede/mídia social. A coleta dos dados foi realizada usando a Aplication Programming Interface do Youtube, o pré-processamento usando bibliotecas do Python e a análise dos comentários utilizando um léxico adaptado para o português brasileiro denominado "Léxico para Inferência Adaptada". Os resultados apresentaram que as opiniões dos torcedores com relação ao técnico de futebol e a performance do clube no campeonato foram majoritariamente negativas. Conclui que o objetivo proposto foi contemplado e que com o grande volume de dados presentes nos canais oficiais dos clubes de futebol, é possível mensurar a opinião dos torcedores no Youtube para auxiliar os clubes no êxito esportivo. Apresenta como proposta de trabalhos futuros a criação de um modelo baseado nas curtidas dos comentários dos vídeos e a possibilidade da automatização da coleta dos comentários dos vídeos do Youtube. Abstract: Youtube is one of the social networks most used by Brazilians and the potential for content dissemination through videos has led Brazilian soccer clubs to give greater attention to the platform in order to establish a better relationship with their fans. The general objective of this work is to identify and analyze the feelings of the video comments on the YouTube platform, in the context of the performance of the Coritiba Futebol Clube soccer coach, in his early season trajectory in the 2023 Paranaense Soccer Championship. As methodology, the Python language and some supporting libraries were defined as the tool to be used for the collection, pre-processing and analysis processes of the data extracted from the network/social media. The data collection was performed using Youtube's Application Programming Interface, the pre-processing using Python libraries, and the analysis of the comments using a lexicon adapted for Brazilian Portuguese called "Lexicon for Adapted Inference". The results showed that the opinions of the fans regarding the soccer coach and the club's performance in the championship were mostly negative. It concludes that the proposed goal was contemplated and that with the large volume of data present in the official channels of the soccer clubs, it is possible to measure the opinion of the fans on Youtube to help the clubs in the sporting success. Presents as a proposal for future work the creation of a model based on the likes of the video comments and the possibility of automating the collection of Youtube video comments.
Collections
- Gestão da Informação [558]