Research On Short-term Traffic Flow Forecast Of Online Car-hailing Based On Combined Neural Network

In recent years,with the rapid development of urban economy,the rapid growth of the number of cars,the rapid increase of traffic load,the problem of urban traffic congestion has become increasingly serious.The rapid development and application of modern information technologies such as big data,Internet of Things,5G,and artificial intelligence have provided new ideas for improving traffic conditions,and intelligent transportation systems have emerged.Among them,traffic flow prediction is one of the prerequisites and key technologies for the realization of intelligent transportation.This paper takes urban car-hailing big data as the research object,combined with the characteristics of traffic time and space data,and uses a variety of neural network models to predict short-term traffic flow.The main work content is as follows:(1)By studying the principles and network structure of convolutional neural network CNN,long and short-term memory neural network LSTM and attention mechanism,summarize their characteristics and advantages,and select a short-term traffic prediction model suitable for urban car-hailing big data.(2)Analyze and preprocess the selected city online car-hailing big data set,clean the repeated data,process the abnormal data,and match the GPS track points with the electronic map.At the same time,it uses data visualization technology to analyze the data set,which shows the inner relationship of the data more deeply.(3)Aiming at the problem of short-term flow forecasting,according to the periodic characteristics of traffic flow,three components are used to model the traffic data.Use CNN and LSTM to obtain the temporal and spatial dependence of traffic data,and combine the attention mechanism to dynamically adjust the degree of influence of different historical periods on the target period.Next,the results of the three components are weighted and merged to obtain the final prediction result.And carried out experimental design,analysis and demonstration of experimental results.The experimental results show that the method of using the combined neural network model for online car-hailing short-term traffic flow forecasting proposed in this paper can be better evaluated,and has certain theoretical and practical application value.