rabieifk / uberdemand_recurrent_neural_network

Demand prediction for Uber based on Multivariate Time Series Forecasting with LSTM (long short-term memory)

UberDemand_Recurrent_Neural_Network

Objective: Multivariate Time Series Forecasting with LSTMs for prediction in UberDemand dataset

Forecasting is one of the applications of RNNs (Recurrent Neural Networks), and in this project, the objective is to use an RNN to predict the number of online uber taxi calls in four parts of New York City for the upcoming hour.

UberDemand.csv contains the number of requests per hour for each section, as well as weather and time information, from January 1 to June 30, 2015.

RNN, LSTM and GRU are the networks that can use in prediction.

The RNN in this project was implemented based on the Keras library.