Using Markov Models to Learn the Sentiment of Soccer Fans from Bets and the Result of Matches

In this paper we investigate variations of Hidden Markov Models (HMM) as a viable tool for predicting the sentiment of soccer fans based on information regarding the result of matches. The models were constructed from data collected from a social network where fans of a soccer team periodically express feelings towards their team. Our claim is that the change in a fan´s sentiment is analogous to a Markovian process of change of state through time. A comparative evaluation performed between variations of the proposed models showed that a second order HMM, considering the match results and fan´s gambling information, is the most accurate model.