A data fusion ticket clearing method

Abstract

The application discloses a kind of data fusion's ticket clearing method, reconstructs passenger travel trajectory based on passenger mobile phone signaling data, constructs path selection probability estimation model based on mobile phone signaling data;Based on AFC transaction data, construct passenger path selection probability likelihood function, obtain path selection probability estimation model based on AFC transaction data;Logit model is constructed to estimate the credibility of the path selection probability model estimation result based on mobile phone signaling data and based on AFC data, the path selection probability estimated by the path selection probability model based on mobile phone signaling data and AFC data is weighted summation to obtain the credibility as weight Fusion path selection probability estimation model is trained and tested;According to ticket clearing after training, by constructing data fusion's urban rail transit path selection probability estimation model, simultaneously adapt to the passenger flow characteristics of different operation time, different operation area of city rail network, adapt to the accurate clearing of different passenger flow characteristics.

Images