| In the competitive sports, sports injury serious impact on the athlete’s technical level, become one of the main reasons for the athletes eliminated. Sports injuries affect normal race and training, and once it happens, it’s very hard for recovery, and may even shorten life, cause physical disabilities, and cause great damage to the state. In the sports training, the sports injury caused by many factors, such as athletes’ physical qualities, health, the weather, equipment and so on. If we can find the most important factors and the potential relationship between factors and sport injury, and control in the later sport training, which will greatly reduce the happening of the sports injury.With the development of the multi-discipline, all kinds of advanced technology into the sports field. Rough set theory as a new mathematical tool which based on the classification capability, can find internal relationship or decision rules by the reduction of knowledge, and not need any prior knowledge, so it has been applied widely in various fields. By studying the attribute reduction algorithm of rough set theory, the paper will realize a reasonable and effective reduction algorithm, which used to dig out the main risk index, and then set up a early warning model of track and field injuries based on attribute reduction algorithm.By studying all kinds of algorithms, combining with the current situation of track and field athletes in Shandong, analyzing current reduction algorithm, the paper put forward and realize two reduction algorithms based on attribute significance. At last, we got two results. Then a BP neural network model was established, simulation tests were carried out with training samples proved by80%of sport athletes. By comparing the convergence, results and error, confirming the final reduction algorithm and the best model structure. In the end, the model is simulated under MATLAB with the20%of the sample as test data, and got the satisfactory results. |
PDF Full Text Request