matlab code for individual metro trip destination inference based on smart card data
Each user contains a set of tuples (origin, time, destination) {(w^o, w^t, w^d)}. The objective is to get the destination probability of a new trip p_u(w^di|w^oi,w^ti,w^o{all except i},w^t{all except i},w^d{all except i}).
graphical model
require mex
mex mnrnd_mex_noscale.c
main.m
input data (train/test) should contain the following entries:
user_id,station_x,station_y,hour_x,hour_y
Reference:
Cheng, Z., Trépanier, M. & Sun, L. Probabilistic model for destination inference and travel pattern mining from smart card data. Transportation (2020). https://doi.org/10.1007/s11116-020-10120-0