Identifying travel direction of cellular phone using Naive Bayes

There are several advantages of using cellular network information for traffic collection and personal navigation including its widely available infrastructure and high penetrate rate, especially in metropolitan area. For such applications, travel direction is a significant parameter. In this study, we proposed a novel method to identify the travel direction of cellular phones based on the handoff information from cellular network. The focus of this research is to identify the travel direction of cellular phone based on 2 attributes, i.e., (1) number of intersection member (IM) and (2) accumulated number of intersected member (AIM) using Naive Bayes model. The experimental results show promising performance with accuracy up to 100%. This suggests a step forward to the use of cellular-based information as road traffic information.