Football players performance analysis and formal/informal media: Sentiment analysis and semantic similarity

• Autor(a) principal: Silva, Gustavo Henrique de Sousa
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/24132
• topic: Football; Work domain analysis; Match annotation; Performance analysis; Sentiment analysis; Futebol; Análise de performance; Análise semântica -- Semantic analysis; Análise de sentimento

Football teams, not different from what already occurs in other kinds of activities in the modern sports world, try to obtain, as much as possible, consistent information on athletes. Currently, it is very important to control not only the physiology, nutrition, and health indicators, but several other aspects. Player’s performance can be measured in an objective way (e.g., Goals scored, assists, interceptions), this being seldom a method to compare and rank the best players by categories. Over years of study, many other factors that can influence the players performance have been discovered and studied, considering not only objective factors, but also subjective factors. Match commentary from different sources (e.g., social and formal media) also plays an important role on a more subjective performance assessment. By using semantic similarity analysis this study aims to contribute to the understanding of the concepts that are used in commentaries, notably key phrases associated to match processes used in entries published in social and formal media. This work also aims to analyse the sentiment about teams, players, and coaches in social media, thus, it explores the fans’ and specialised media perspective on performance. By using named entity recognition and sentiment analysis tools over the set of comments and opinions expressed by fans on Reddit and specialized media on sports sites about a UEFA Champions League match it was possible to distinguish different sentiments on different entities (players, coaches, teams) and relate that with objective aspects of that match. In addition to these results, it was also possible to identify several shortcomings of the usage of these tools in this context (e.g., usage of irony, slang). The obtained results showed the comments of fans and formal media about a match as presented in the Similarity Analysis of the Work Domain Analysis (WDA) structure. Through the Sentiment Analysis it was possible to describe some impressions about the most commented topics and the relation with some match events.