Spatiotemporal Traffic Flow Prediction with KNN and LSTM

Abstract

The traffic flow prediction is becoming increasingly crucial in Intelligent Transportation Systems. Accurate prediction result is the precondition of traffic guidance, management, and control. To improve the prediction accuracy, a spatiotemporal traffic flow prediction method is proposed combined with k-nearest neighbor (KNN) and long short-term memory network (LSTM), which is called KNN-LSTM model in this paper. KNN is used to select mostly related neighboring stations with the test station and capture spatial features of traffic flow. LSTM is utilized to mine temporal variability of traffic flow, and a two-layer LSTM network is applied to predict traffic flow respectively in selected stations. The final prediction results are obtained by result-level fusion with rank-exponent weighting method. The prediction performance is evaluated with real-time traffic flow data provided by the Transportation Research Data Lab (TDRL) at the University of Minnesota Duluth (UMD) Data Center. Experimental results indicate that the proposed model can achieve a better performance compared with well-known prediction models including autoregressive integrated moving average (ARIMA), support vector regression (SVR), wavelet neural network (WNN), deep belief networks combined with support vector regression (DBN-SVR), and LSTM models, and the proposed model can achieve on average 12.59% accuracy improvement.

Evaluation of Short-Term Freeway Speed Prediction Based on Periodic Analysis Using Statistical Models and Machine Learning Models

“…Unlike VAR model, ARIMA model only takes the e ect of time series into account, and has been used in various tra c data analysis [60][61][62]. A nonseasonal ARIMA model can be de ned as follows: ARIMA ( , , ), where, is the number of autoregressive terms, is the number of non-seasonal di erences and is the number of lagged prediction errors.…”

“…SVC, addressing the classi cation problems, calculates a decision boundary and maximizes the distance between the boundary and the nearest sample data. Like SVC, SVR uses a similar approach for regression problems and ignores the error which is less than between the observed value and the estimated value [60,63,64]. More speci cally, given a group of training data, the objective is to seek a function ( ) that the maximum deviation between actual values and predicted values is at most .…”

Influence of Ground Temperature on Shotcrete-to-Rock Adhesion in Tunnels

“…e terrestrial heat in the construction environment not only worsens the working conditions of mechanical equipment but also increases the number of failures; it also seriously jeopardizes the health and safety of workers. erefore, it is necessary to analyse the causes of high ground temperature in combination with actual engineering examples (Table 1) and improve the understanding of the high ground temperature environment, so as to better resolve the hazards [15,16,[19][20][21][22].…”

Dynamic Failure Mode and Dynamic Response of High Slope Using Shaking Table Test

“…e area hosts typical landslide-prone strata. Earthquake landslides have become the most common geological hazard in the construction of these areas [1][2][3][4]. Earthquake landslides are extremely dangerous and have caused severe casualties and property losses worldwide.…”