摘要
短时交通流的准确高效预测对于智能交通系统的应用十分关键,但较强的非线性和噪声干扰使其对模型的灵活性要求较高,并且还需在尽可能短的时间内处理大量的数据。因此,讨论了用随机森林模型对短时交通流进行预测,该模型具有比单棵树更强的泛化能力,参数调节方便,计算高效,且稳定性好。观察交通流数据在较长时间跨度上的变化后,提取出主要特征变量构造输入空间,对模型进行训练后,在测试集上的预测准确率约为94%。与目前广泛使用的支持向量机模型进行对比分析,结果显示随机森林预测不仅准确率稍好于支持向量机,而且在效率、易用性及未来应用的扩展上都要优于支持向量机。
The short term traffic flow prediction is very important to the application of intelligent traffic system（ ITS）,but it needs more flexible model for the strong nonlinear and noisy and processes lots of data in short time. This article discusses the random forest model for the prediction of short term traffic flow. The model has characters such as stronger generalization,easy to adjust the parameter,effective computation and quality stability. It extracts the principal features as the variables to form input space after observing the variation of traffic flow in the longer term. The prediction accuracy of the model on the test set is 94% after the model trained on the training set. Compared with the popular support vector machine（ SVM）,the random forest has better accuracy prediction. And the random forest is better than SVM on the efficiency,usability and the extension of future usage.
出处
《微型机与应用》
2016年第10期46-49,共4页
Microcomputer & Its Applications
关键词
智能交通
交通流预测
决策树
随机森林
支持向量机
intelligent traffic system
traffic flow prediction
decision tree
random forest
support vector machine
