With the continuous development of society, the cooperation of different dimensions is urgently needed. Analysis and modeling of team cooperation model and performance evaluation are especially important for competitive sport. In this paper, a football team’s attacking mode and the team performance were assessed using network science methodologies. The match process was analyzed by using the data of Team A (given in the form of attachment due to excessive file size) and the method of complex network science. Each player was regarded as a node in the network, and the interaction among players was considered as the connection to the network. This method directly reflected the favorable formation of the team and the interaction frequency among members. Then, a team performance evaluation model was established using the backpropagation neural network (BPNN) and the uncrossed analytic hierarchy process (U-AHP) method based on the factors including the number of passes and successful pass rate. The team performance was comprehensively rated from two levels: member and team level. Analysis from established models indicated that Team A had a higher probability of winning when using the “4-4-2” offensive strategy and performance evaluation analysis indicated that more passes and higher pass success rates were more beneficial to win the game. Following the model developed in this study, some suggestions were given from the perspectives of team strategy, attack mode, cooperation, and incentive mode.

中文翻译：

---

基于网络科学和BP神经网络的足球队协作模式与绩效评估分析与建模

随着社会的不断发展，迫切需要不同层面的合作。团队合作模型的分析和建模以及绩效评估对于竞技体育尤为重要。在本文中，使用网络科学方法评估了足球队的进攻方式和球队表现。通过使用团队A的数据（由于文件过大而以附件形式给出）和复杂的网络科学方法来分析比赛过程。每个玩家都被视为网络中的一个节点，玩家之间的交互被视为与网络的连接。这种方法直接反映了团队的有利组成和成员之间的互动频率。然后，基于通过次数和成功通过率等因素，使用反向传播神经网络（BPNN）和不交叉分析层次过程（U-AHP）方法建立了团队绩效评估模型。团队绩效从两个级别进行了综合评分：成员级别和团队级别。对既有模型的分析表明，使用“ 4-4-2”进攻策略时，甲队获胜的可能性更高，而性能评估分析表明，越多的传球和更高的传球成功率对赢得比赛更有利。根据本研究开发的模型，从团队策略，攻击模式，合作和激励模式的角度提出了一些建议。