提出一种基于MapReduce的聚类和神经网络相结合的公交车到站时间预测模型.首先,结合公交车的运行特点,利用K-means聚类方法对公交车的运行时段进行划分,同一时段中的公交车运行数据具有较高的相似性;然后,分别对各个时段的公交车运行数据建立BP神经网络模型进行到站时间的预测;其次,在大数据平台上,针对聚类和神经网络相结合的分段预测模型,建立了基于MapReduce的并行化框架.最后,以公交车的实际运行数据为例进行仿真与验证.实验结果表明,该分段模型优于传统的BP神经网络预测模型,具有较高的预测精度和预测速度.%Bus arrival time prediction is the foundation of promoting intelligent bus service.Aiming at the prediction problem,a model of predicting bus arrival time combining MapReduce clustering with neural network was proposed.First of all,K-means clustering method was used to divide time-interval of the bus according to the phenomenon that the bus in the same time-interval has a high similarity.Secondly,according to the data of different time-intervals,BP neural network was established respectively to predict bus arrival time.Thirdly,for the section-model of predicting bus arrival time,the parallel framework based on MapReduce was established.And lastly,the Simulation and verification was made on the actual bus operating data.The experimental results show that the proposed model is superior to the traditional BP neural network with higher prediction accuracy and speed.
展开▼