Traffic fluctuation prediction device, method and program

Abstract

The present disclosure has been made in view of such a problem, and an object of the present disclosure is to make it possible to predict fluctuation of unsteady traffic with a small amount of calculation. A traffic fluctuation prediction apparatus () according to the present disclosure includes: a data division unit () that divides time-series data X(t) in a certain period of the traffic into estimation and prediction data sets; a learning unit () that learns a dictionary D(t) using one of the two divided data sets, and learns a dictionary D(t) using the other of the two divided data sets; a prediction unit () that obtains a sparse code Y(t) in representing the time-series data using the learned dictionary D(t) and obtains a predicted value X{circumflex over ( )}(t) of future traffic using the obtained sparse code Y(t) and the dictionary D(t); and a prediction correction unit () that corrects a prediction error of the predicted value X{circumflex over ( )}(t) obtained by the prediction unit () using search for a stable region of a double queue of overestimation and underestimation.