基于决策树算法的篮球联赛预测

摘要这些年来，体育竞技越来越受到人们的关注，而体育竞技的分析以及预测比赛结果也已经演变成为一个重要的领域。为了提高预测竞技比赛的正确率，本文以 NBA 篮球联赛为例，基于决策树算法，通过提取不同的特征，对预测结果的准确度进行不断的提升，并找出最优算法。89129

在研究过程中由于影响球队胜负的因素很多，所以需要抽取大量特征进行计算，并最好一次性处理所有特征。单个决策树在许多方面可能会导致过拟合问题，因此，需要调整决策树算法——限制其规则数量。最终得到预测结果的正确率相对最高的算法为决策树算法的优化——随机森林。

Over the years, sports have received more and more attention, and the analysis of sports and the results of the competition have also evolved into an important area。 In order to improve the accuracy of the competitive competition, this paper takes the NBA Basketball League as an example, and based on the decision tree algorithm, the accuracy of the prediction results is improved and the optimal algorithm is found by extracting different features。源Y于U优I尔O论P文W网wwW.yOueRw.com 原文+QQ75201-8766

In the course of the study because of the factors affecting the outcome of the team a lot, so the need to extract a large number of features to calculate, and the best one: time treatment of all features。 A single decision tree may lead to a fitting problem in many ways, so it is necessary to adjust the decision tree algorithm to limit the number of rules。 Finally, the optimal rate of the predicted results is the highest algorithm for the optimization of the decision tree algorithm: random forest。

毕业论文关键词：决策树; 预测; 随机森林; pandas 库

Keyword: decision trees; forecast; Random forests; pandas library

摘要 2

1。绪论 4

1。1。问题背景 4

1。2。问题意义 4

2。 NBA 的网络资源 5

3。数据处理 6

3。1。载入数据 6

3。1。1。收集来自优Q尔W论E文R网wWw.YouERw.com 加QQ75201.8766 数据 6

3。1。2。使用pandas 来加载数据 6

3。1。3。提取新特征 7

3。2。决策树 9

3。2。1。决策树中的参数 9

3。2。2。决策树的使用 10

3。3。体育成果预测 10

3。4。随机森林 14

3。4。1。系统如何工作 14

3。4。2。随机森林的参数 15

3。4。3。随机森林的应用基于决策树算法的篮球联赛预测:http://www.youerw.com/shuxue/lunwen_177226.html