Search

Food ball is a prevalent sport in the world, based number of people interested, the number of leagues per year, and the economy depended on it. So, predicting and analyzing the game is very challenging. The performance depends on many parameters like players, defense strength, ground weather, toss, and previous team and individual performance. So many researchers have worked on predicting the result, but most researchers used simple machine learning models, tried to increase the accuracy, and missed some essential features. Moreover, football is continuing game; if a player is not playing or his performance is poor in the last match, that will affect future results. So, we proposed a new model with Bi-LSTM to learn the team's and individual players' performance within the period. Though we got less accuracy than other models, our model precision, recall and F-score are up to the mark. [ABSTRACT FROM AUTHOR]