Predictive Model of Personalization of Services of Automated Mobility Based on the Records of User Movement in Mobile Networks

Abstract

The aim of this paper is to investigate and develop a predictive model of personalization of automated mobility services based on historical and realtime data on the trajectory of movement of users in mobile networks. The problem is based on the fact that the concept of automated mobility enables a variant of modeling the parameters of the vehicle trajectory and personalization of the service according to the wishes and needs of users, whereby optimization and personalization can be performed both on the user side and on the side of the mobility service provider. The paper uses a research approach within the domain of intelligent transport systems based on the extraction of telecommunication activity data into the index of personalization of mobility parameters using the methodology of predictive analytics and neural inference based on neural network. The research findings establish and show a clear correlation of historical and realtime spatial data of the movement of users and their activities in the mobile network with indicators of quality of urban and sustainable mobility. The significance of the model is that only a satisfied user of a personalized automated mobility service no doubt intensely and more frequently uses a given service on the one hand, and the automated mobility system can predictively extract a set of parameter values that satisfy that user and offer the optimum variant of the service.