Neural network prediction model research of shuttlecock competition based on data mining technology

In the sports game, it is essential if we can predict the results of the competition. This paper first analyzes the status of a shuttlecock team using data mining technology and predicts the team's mathematical expectation and development trends. On this basis, it uses the homogeneous Markov model to validate the future development trends of the game, and gets a reasonable conclusion that the state of a good shuttlecock team is stable. Secondly, starting from the shuttlecock athletes' athletic ability, it constructs a neural network fuzzy comprehensive evaluation model of athlete's athletic ability. As long as the transition probability matrix is non-diagonal matrix, regardless of the first game is won or lost, it will not affect the back game's achievements, and provides a more reliable and scientific approach and methodology for the evaluation of athletic ability. © 2013 by CESER Publications.