A user-personalized model for real time destination and route prediction

• Published in: 2016 IEEE 19th International Conference on Intelligent Transportation Systems (ITSC) pp. 401 - 407
• Main Authors: Dantas Nobre Neto, Francisco, De Souza Baptista, Claudio, Campelo, Claudio E.C.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.11.2016
• Subjects: Context; Global Positioning System; Hidden Markov models; Markov processes; Predictive models; Roads; Semantics
• ISSN: 2153-0017
• DOI: 10.1109/ITSC.2016.7795586

The act of predicting a destination and route a user will take, as soon as he/she begins to move, has several benefits. A system with this kind of information is able to help the user to avoid a congested route or to suggest a Place of Interest (POI). Nowadays, the task of tracking a user movement is more feasible thanks to current smartphones, with embedded GPS devices. Many related work addresses the problem considering only next place prediction. They do not focus on route and destination prediction. Moreover, many models for place prediction only predict a place already visited by the user. This paper proposes a user-personalized model for predicting route and destination, including places where user has never visited before. Our model works automatically, i.e., without user interactions. The experiment was conducted with a real and public dataset, containing data from 21 users collected during three months, and the results were encouraging, with a precision of 76.9%.